Morning Exam Section 1 Basic Theory Answers and Explanations

Section 1 Basic Theory

Q1-1 c) Representation of a hexadecimal fraction as a decimal fraction)

If a hexadecimal fraction is represented in decimal, the result is typically as shown below.

(…x 2 x 1 x 0 . y 1 y 2 y 3 …)16
2 1 0 1 2 3
=…+x2×16 +x1×16 +x0×16 +y1×16− +y2×16− +y3×16− +…

If the hexadecimal fraction 0.C is represented in decimal and calculated as below, the answer is
c) 0.75.
−1 1 12 3
0.C = (C)16×16 = 12× = = = 0.75
16 16 4

Q1-2 a) Representation of a fraction in a radix conversion

This question deals with binary, octal, and decimal values. First, concerning binary and octal
3
values, 2 =8 means that a binary value can be split into blocks of three digits and substituted for an
octal value. This also means that the notations for octal and binary values are compatible. For this
reason, in a radix conversion between a binary value and an octal value, there are no phenomena
where a finite fraction becomes an infinite fraction or vice versa. Thus, whether or not the result of
a radix conversion becomes a finite fraction or an infinite fraction only needs to be considered
between a decimal value and a binary value (octal value).
(1) When a decimal value with a finite fraction is represented in binary value
The examples of a finite fraction and an infinite fraction are clearly shown below.
[Example]
• (0.5)10=(0.1)2: Example of a finite fraction
• (0.1)10=(0.00011001100110011…)2: Example of an infinite fraction
(2) When a binary value with a finite fraction is represented in decimal value
The fraction is represented as shown below.

1 1 1 1 1
a1 + a2 2 + a3 3 + ..... + an −1 n −1 + n
2 2 2 2 2
n
1
2 ∑a
k =1
k
2k
(ak = 0 or 1)

1
is a finite fraction, so the previous expression is the sum of a finite fraction and its
2k
value is also a finite fraction. Thus, if a binary value with a finite fraction is represented in
decimal value, the result is a finite fraction.

1
 Morning Exam Section 1 Basic Theory Answers and Explanations

a) (2) shows that this is correct.

b) This is incorrect, as explained above.
c) Octal values and binary values are equal, so this contradicts (2) and is incorrect.
d) (1) shows that this is incorrect for binary and so is also incorrect for octal.

Q1-3 a) Method for checking if a binary value is a multiple of 16

For an 8-bit unsigned value, the weighting is

7 6 5 4 3 2 1 0
(2 2 2 2 2 2 2 2 ) = (128 64 32 16 8 4 2 1)
from the left digit. Each of the upper four digits is a multiple of 16, so no matter which bit is one,
it is a multiple of 16. The weighting of each of the lower four digits is a number of 8 or less, and if
all digits are aggregated the result is 15, so if any bit is 1 then division by 16 is not possible. Thus,
if the lower four bits of a binary value are all zeros, then it is a multiple of 16.
In order to retrieve the lower four bits of x, the logical sum of x and a binary value with 1 in the
lower four bits should be calculated. Therefore, a) is appropriate.

Q1-4 c) Register value after a shift

When overflow does not occur and a binary value is shifted one bit to the left, the value
1 2
becomes 2 =2 times the original value. If the shift is two bits, the value becomes 2 =4 times the
3
original value, and if the shift is three bits, it becomes 2 =8 times the original value, and so fourth.
This is the same as multiplication by the power of two, so if the shift is two bits to the left the
value becomes four times the original value. In the conditions of the question, the original number
x is added to this value. If x is added to the original value multiplied by four (4x), it becomes 5x,
which shows that the value is the original value multiplied by five. Therefore, c) is correct.
a) If x is shifted one bit to the left, it becomes 2x, and if x is then added, it becomes 3x.
b) If x is shifted two bits to the left, it becomes 4x.
d) If the result (i.e., 2x) of shifting x one bit to the left is added to the result (i.e., 4x) of shifting
x two bits to the left, it becomes 6x.

2
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-5 c) 2’s complement

The 2’s complement of a binary number is the value obtained by subtracting the binary number
from a reference value (i.e., “the maximum value possible with the same number of digits as the
number for which 2’s complement is to be calculated” +1). When a negative number is to be
represented with 2’s complement, the 2’s complement of the absolute value of that number is used.
In this question, the number of digits is eight, so the result of subtracting the absolute value from
100000000 (the maximum value of eight digits +1) is 10101110. If the absolute value is
represented as x, the relationship is as show below.
100000000 − x = 10101110 (← this is the 2’s complement of x)
x = 100000000 − 10101110
x = 01010010
Therefore, c) is correct. 2’s complement can also be calculated by inverting the bits and adding 1.
In the same way, if the reverse operation is performed, 1 is subtracted and the bits are inverted:
10101110 − 1 = 10101101 → inversion: 01010010). As a result, the same value can be obtained.

3
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-6 c) Range represented in a fixed point format

In a fixed point format that represents negative numbers with 2’s complement, a positive or
negative sign is identified as the leftmost bit. 0 is for a positive number and 1 is for a negative
number (when the number of 0 is represented, all bits are 0s but this is not a positive number). The
maximum positive number that can be represented with n bits has “n −1” digits.
When a negative number is represented with a 4-bit fixed point number, the result is shown in
the table below. An examination of the leading one bit of each binary value in the table shows that
it is 0 for a number of 0 or greater and 1 for negative numbers. In order to apply this law, (1000)2
is regarded as − 8, not +8.

Decimal value Binary value

+7 ---------- 0111
+6 ---------- 0110 Calculated with
+5 ---------- 0101 2’s complement
+4 ---------- 0100
+3 ---------- 0011
+2 ---------- 0010
+1 ---------- 0001
0 ---------- 0000
−1 ---------- 1111
−2 ---------- 1110
−3 ---------- 1101
−4 ---------- 1100
−5 ---------- 1011
−6 ---------- 1010
−7 ---------- 1001
−8 ---------- 1000

Based on the above table, if a fixed point number with four bits is considered, (− 8)10 through
3 3
(+7)10 can be represented. This range can also be represented as (−2 )10 to (+2 − 1)10. The table
below shows the representable range for each length of a fixed point number.

4
 Morning Exam Section 1 Basic Theory Answers and Explanations

Length of fixed Representable range Alternative Type of

point number representation representable number
3 3 4
4-bit − 8 through +7 − 2 through +2 − 1 2 (=16)
7 7 8
8-bit − 128 through +127 − 2 through +2 − 1 2
15 15 16
16-bit − 32,768 through +32,767 − 2 through +2 − 1 2
31 31 32
32-bit ………………… − 2 through +2 − 1 2
n−1 n−1 n
n-bit ………………… − 2 through +2 − 1 2
As such, with a fixed point number of n bits in length, the values that can be represented are
n−1 n−1
− 2 through +2 − 1. Therefore, c) is correct.

Q1-7 d) Conversation of decimal values to binary values

This question considers the conversion of decimal integers to unsigned 1-byte binary integers.
a) (120)10 = (01111000)2
b) (127)10 = (01111111)2
c) (170)10 = (10101010)2
d) (240)10 = (11110000)2
When b) 127 is converted to a binary integer, the number of zero (0) bits differs from that of one
(1) bits, so it not a possible answer. Thus, the binary integer with the maximum value should be
chosen from a), c), and d). Therefore, d) is correct.

Q1-8 a) Floating point representation of a real number by using an expression

When the real number a is represented as a = x × y z with floating point representation system,
the name of each part is as shown below. Therefore, a) is appropriate.
Sign Fraction Radix Exponent

a = ±x × yz

A floating point number that is represented in a data format in a computer is as shown in the
figure below.

Exponent Fraction

▲
Sign bit Location of radix
point

5
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-9 c) Errors in floating point arithmetic operations

The answer that applies to the question is c) loss of trailing digits. In the addition and
subtraction of floating point numbers, before an operation, the values of the exponents of two
numbers are adjusted to the appropriate size. In other words, the value of the exponent of the
number with the smaller absolute value is adjusted to the value of the exponent of the larger value,
and the fraction part of the smaller value becomes zero, which is the cause of the phenomenon. In
the addition and subtraction of multiple values, in order to prevent the loss of trailing digits,
addition or subtraction is performed after the values are arranged in order of absolute value from
smallest to largest. By doing this, the operation is on absolute values that have a small difference,
and the loss of trailing digits is unlikely to occur.
a) Truncation error is an error that occurs in an operation such as an engineering calculation
where an infinite fraction is cut off.
b) Cancellation of significant digits is a phenomenon where the number of valid digits
decreases due to normalization of the fraction when two numeric values with the same sign
bit and almost the same absolute value are added or subtracted.
d) Absolute error is the actual value of an error. The ratio of an error to the size of a value is
called a relative error.

Q1-10 c) Relationships that are always valid in sets

The Venn diagram below is considered here.

A B (4)

(1) (2) (3)

If sets are represented with (1) through (4) in the above figure so that A is (1) and (2), set B is
(2)and (3), and A∩B is (2), then each option can be replaced as shown below:
a) A ⊆ ( A  B ) → (1)(2) ⊆ (1)
(1)(2) ⊆ (1) means “(1) and (2) is a subset of (1).” A subset means that the elements of a
given set are completely contained in another set. In this case, (1) and (2) is not completely
contained in (1), so it is not valid.
b) ( A  B ) ⊆ ( A  B ) → (1)(2)(3) ⊆ (1)(3)(4) … Not valid
c) ( A  B ) ⊆ ( A  B ) → (2) ⊆ (1)(2)(4) … Valid
d) ( A  B ) ⊆ ( A  B ) → (2) ⊆ (4) … Not valid
Therefore, c) is correct.

6
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-11 a) Equal values in logical expressions

Based on De Morgan’s laws:

A• B = A+ B
A+ B = A• B,
the given logical expression can be broken down as shown below:
( A + B) • ( A + C ) = ( A + B) + ( A + C )
= ( A • B) + ( A • C )
= A• B + A•C
Therefore, a) is correct.

Q1-12 b) Logical expressions for functions shown in a truth table

In this kind of question, confirmation of each option can ultimately lead to a faster answer. A
truth table for each option can be written and considered, but attention should be paid to the fact
that each value within the parentheses of the options is a logical product (true only when both
values are true) and the parentheses are linked with logical sums (true if at least one is true). Then,
an appropriate value should be entered into the variable, and a comparison with the result of
f (x, y, z) of the truth table should be performed.
First, the first half of a) to c) is (x ∧ y), so if the values x = T, y = F, and z = T of the third row
of the truth table are inserted, the result is x ∧ y = F.
a) In the latter half, y ∧ z = F, so the overall result is also F. This does not match the result of
f (x, y, z) in the truth table.
b) In the latter half, － y ∧ z = T, so the overall result is also T. This matches the result of
f (x, y, z) in the truth table. Therefore, b) is correct.
c) In the latter half, － y ∧－ z = F, so the overall result is also F. This does not match the result
of f (x, y, z) in the truth table.
d) Only for this option, the values x = T, y = T, and z = F of the second row which make the
result of the first half F are inserted. x ∧ － y = F, － y ∧－ z = F, so the overall result is also F.
This does not match the result of f (x, y, z) in the truth table.
The answer is understood to be b), but other values should be inserted for confirmation before
finally providing an answer.

7
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-13 d) The shortest path viewed from a diagram

A question where the number of shortest paths is calculated can be considered as a question
about permutations and combinations. The shortest path from P to R is a total of four steps
composed of two horizontal and two vertical steps. The path depends on the order of these steps,
so the number of combinations of these steps is the number of shortest paths. This can be
calculated as a combination that selects two from four.
P→R: 4C2=4×3/(2×1)=6
In the same way, the number of shortest paths from R to Q is decided by the order in which two
horizontal and three vertical steps are taken. This can be calculated as a combination that selects
two from five.
R→Q: 5C2=5×4/(2×1)=10
Here, when P is to be reached from Q, the point R must be passed. Thus, the number of the
shortest paths can be calculated by multiplication (independent of each other) of the combinations
for P to R and R to Q. The answer is d) 6×10=60 paths.

Q1-14 d) Calculation of probability

Probability represents the degree of certainty that a given event will occur as a numerical value.
It can be calculated with the expression described below:

Number of times a given event can occur

Probability that a given event will occur =
Total number of events that can occur
When one ball is removed from a bag that contains four white balls and five red balls and then
another ball is removed without putting the first ball back into the bag, the probability that both
balls are red is as shown below:
• Probability that the first ball is red
The number of balls in the bag is nine, of which five are red. Thus, the probability that the
first ball is red is 5/9.
• Probability that the second ball is red
The first ball is not put back into the bag after it is removed, so the number of balls that
remain in the bag is eight. As it is assumed that the first ball removed is red, the number of red
balls that remain is four. Thus, the probability is 4/8 = 1/2.
The probability that both balls are red is calculated from the product of both probabilities:
(5/9)×(1/2) = 5/18. Therefore, d) is correct.

8
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-15 d) Calculation of the probability of a change in the weather

A state transition diagram is a diagram that is used to represent a transfer within a finite number
of states when the next state is decided only by the current state and an incoming event. State
transition diagrams were originally used in the design of sequential circuits for computers.
However, as they have the versatility to be used for things such as analysis of grammar in
programming languages and task control in operating systems, they can also be used for the
description of logic.
For the state transition in this question, when the weather is rainy on a given day, there are three
cases in which the weather is clear two days later.
(1) Probability of day 1 rainy → day 2 rainy → day 3 clear = 0.2×0.3 = 0.06
(2) Probability of day 1 rainy → day 2 clear → day 3 clear = 0.3×0.4 = 0.12
(3) Probability of day 1 rainy → day 2 cloudy → day 3 clear = 0.5×0.3 = 0.15
For the probability that is calculated, any of the cases from (1) to (3) can occur, so the sum of
the probability for each case is used for calculation. Therefore, the result is 0.06+0.12+0.15=0.33,
and d) is correct.

Q1-16 b) Variation in character strings

The number of character strings with one character that can be expressed using A and B are the
two variations A and B. In the same way, the number of variations for a two-character string is
four: AA, AB, BA, BB. The characters that are used are the two characters A and B, so this
question can be considered in the same way as binary. In the case of binary, for one digit there are
1 2 n
2 variations, for two digits there are 2 variations, and for n digits there are 2 variations that can
be represented. In this question, the number of variations that are possible with a length of one to
seven characters is considered, and the answer is as follows:
1 2 3 4 5 6 7
2 +2 +2 +2 +2 +2 +2 = 2+4+8+16+32+64+128 = 254
Therefore, b) is correct.

9
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-17 a) Graph of normal distribution

A graph that represents normal distribution resembles a symmetric mountain with sides that
smoothly approach the horizontal axis. Standard deviation is a statistical index that represents how
far each item of data is from the average. In a graph for normal distribution, the relationship
between the average and the standard deviation is shown in the diagram below. Therefore, a) is the
correct graph. In quality management, as per the three sigma (3σ) rule, if the standard deviation is
multiplied by three, most data is within that range, and thus it can be used to manage the degree of
variation in quality.

μ－3σ μ－2σ μ－σ μ μ＋σ μ＋2σ μ＋3σ

Line of normal distribution (μ is average, σ is standard deviation)

b) Standard deviation shows the distance between each item of data and the central average,
so this is incorrect.
c), d) The line is not symmetrical, so this cannot be normal distribution.

Q1-18 d) Operations to round average values to the nearest integer

For N observed values, the average value of the sum S can be calculated with S/N, but this
question is about the processing method to round the first decimal place of a value to the nearest
integer.
[ ] is called a Gauss symbol and represents an integer that does not exceed the value in the square
brackets, so the result is an integer value with the fraction rounded. Rounding is an action that
rounds up a digit that is 5 or greater and rounds down a digit that is 4 or less. As a technique for
calculating numeric values, there is a method that adds 5 to a digit and if that digit is 5 or greater it
is forcefully rounded up. By applying this method, for rounding to the nearest integer, S/N＋0.5
S
should be calculated and the digits(s) after the fraction should be discarded. Therefore, d) N + 0.5
is correct.
In the kind of example below, it is very important to confirm that correct rounding can be
performed.
[Example] 3.4……3.4+0.5=3.9 The fraction is discarded to leave 3
3.5……3.5+0.5=4.0 The fraction is discarded to leave 4

10
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-19 c) Sample correlation coefficient

A correlation coefficient is a numeric value that shows the level of association between two items.
The possible range of this value is from −1 to 1. The closer the value gets to 1 the stronger the positive
correlation, and the closer the value gets to −1 the stronger the negative correlation. When the
correlation coefficient is −0.9, the options are graph c) or d) which represents a negative correlation.
−0.9 is a stronger negative correlation than −0.7 and the stronger the correlation the closer the
distribution is to the regression line. Therefore, c) is correct.

Q1-20 a) Inference mechanism of a knowledge base

A system that uses a knowledge base in a given field and can make inferences close to those
made by experts in that field is called an expert system. Therefore, a) is correct.
b) A neural network is a biological term that represents the mechanism of a brain of a living
organism such as a human, but it is also used for a computer system that is modeled on this.
For this reason, there are cases where the latter is definitely called an artificial neural
network. Almost all current computers are designed with a so called Neumann-type
architecture. These computers follow a procedure where each instruction is read from a main
memory, execute each instruction, and store the result in the main memory. However, in a
neural network, if the same structure as the brain of a biological organism is simulated, it is
hoped that the same thought as a biological organism will be possible.
c) Virtual reality is a technological field that aims to give realistic experiences to humans
through realistic stimulation of the five senses such as sight and hearing. This has moved into
the fields of games and training.
d) A fuzzy computer is a computer or software with a mechanism for the representation with
fuzzy logic of information that has unclear boundaries such that a fuel tank has “lots of fuel,”
“not much fuel,” or “quite a lot of fuel.”

11
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-21 c) Reverse Polish notation

Reverse Polish notation (postfix notation) is a method for the notation of an arithmetic
expression in a computer, and it describes an operator after the subject of an operation. For
example, “(X+Y)×Z” means that “X” is added to “Y” and then the result (“XY+”) is multiplied by
“Z”, therefore it is represented “XY+Z×”. The order of the operators is considered, and the
notation is made without parentheses.
In the given “EF−G÷CD−AB+÷+”, “E” and “F” are the subjects of the first operator “−”, so this
means “E− F”. The subjects for the second operator “÷” are the result of “E−F” and “G”, so
parentheses are added in line with the operation priority and the expression becomes “(E− F)÷G”.
Next, the subjects for the third operator “− ” are “C” and “D” so this is “C− D”, and the subjects for
the fourth operator “+” are “A” and “B” so this is “A+B”. Next, the subjects for the fifth operator
“÷” are the result of “C− D” and “A+B” so the expression is “(C− D)÷(A+B)”. The subjects for the
final operator “+” are the result of “(E− F)÷G” and “(C− D)÷(A+B)”, so the expression becomes
“((E− F)÷G)+((C− D)÷(A+B))”. Therefore, c) is correct.

Q1-22 a) BNF and bit strings

As for the basic symbols of BNF (Backus-Naur Form), “::= ” represents that “the left-hand side
is defined as the right-hand side,” and “|” represents “OR.” Elements that are defined are enclosed
in angle brackets, and this can be used for a more complex syntax description.
The content that is defined in this question is considered here. < S > is defined as “01” or
“0 < S > 1”, and in a situation where “0 < S >1” is applied it can also be understood that the
definition of the < S > part is recursive. As such, it can be understood that the following kind of
pattern is allowed:
(1) 01 Result of applying “01” to < S >
(2) 0011 Result of applying (1) to “0 < S > 1”
(3) 000111 Result of applying (2) to “0 < S > 1”
(4) 00001111 Result of applying (3) to “0 < S > 1”
…

In every case, the left-hand 0s of the center and the right-hand 1s are repeated such that the both
numbers are equal. Therefore, a) is correct.

12
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-23 c) Strings that are accepted by a finite automaton

An automaton is a machine that is driven by input values but is also affected by previous states.
Furthermore, a finite automaton is an automaton that has a finite number of states and decides
output and the next state from the current state and input. A finite automaton can be represented
with the type of state transition diagram that is shown in the question. It should begin with the
initial state in the diagram, and should change the state with either 1 or 0. Finally, it should
finish with the accepting state . In each state in the diagram, a symbol is assigned as shown
below. If the state transition of each character string in the options is followed, the state transitions
are as shown in the table below.

a
1 1

0
S b
0 0, 1

0 1
0
E c
1

0 1 0 1 1
a)
S → S → a → S → a → b
0 1 1 1 1
b)
S → S → a → b → E → c
1 0 1 1 1
c)
S → a → S → a → b → E
1 1 1 1 0
d)
S → a → b → E → c → c

Therefore, it can be understood that c) finishes with E.

13
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-24 d) Methods for shortening character strings

The method for representing data, where the same character or the same bit pattern is repeated,
in a short bit string is called run length encoding (i.e., encoding of the length of consecutive data).
For example, when the character “A” is repeated 100 times, by representation of this as “100 As”
rather than “AAA...AAA” the data can be compressed. Especially in monochrome fax data, there
is no information except “black” and “white”, and the same bit pattern is repeated often. As such,
in order to reduce the amount of communication, run length encoding is used. Therefore, d) is
correct.
a) EBCDIC is a character code set, and is mainly used in IBM mainframes.
b) A cyclic code is an error correction code that is attached in order to detect data errors on a
transmission channel.
c) Huffman coding is a method for encoding that was developed in 1952 by David Huffman. In
the same way as run length encoding, Huffman coding is an encoding method that efficiently
shortens character strings and the like. But, in contrast to run length encoding, which is good
for repetitions of the same character or bit pattern, Huffman coding represents frequently
used characters as a short bit string, and represents less frequently used characters as a long
bit string. For example, in English the letters “e,” “t,” and “s” appear frequently but the
letters “q” and “z” do not, so for the former letters a short code is used and for the latter
letters a long code is used. By doing this, the data overall can be compressed. In actual
Huffman coding, encoding is based on the frequency of occurrence of each character in the
subject data.

Q1-25 d) Logical equation that always hold for odd parity

In odd parity, a parity value p is set so that the number of 1s in the data is odd. Thus, if the
number of 1s in d0, d1, …, d7 is odd then p is 0, and if the number 1s is even then p is 1.
The representation of the number of 1s in d0, d1, …, d7, p with the exclusive-OR operation is
considered here. For the odd number of 1s and the even number of 1s, the logical relations are
shown below.
· Odd number: d 0 ⊕ d 1 ⊕ … ⊕ d 7 ⊕ p =1
· Even number: d 0 ⊕ d 1 ⊕ … ⊕ d 7 ⊕ p =0
(Example) 1 ⊕ 1 ⊕ 1 =(1 ⊕ 1 ) ⊕ 1 =0 ⊕ 1 =1, 1 ⊕ 1 ⊕ 0 =(1 ⊕ 1 ) ⊕ 0 =0 ⊕ 0 =0
1 ⊕ 0 ⊕ 0 =(1 ⊕ 0 ) ⊕ 0 =1 ⊕ 0 =1 , etc.
Here, the method for odd parity is used, so the logical equation for the odd number of 1s is valid
and d) is correct.

14
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-26 a) Sequence control

Sequence control is a method for control that starts and completes multiple activities in a
predefined sequence. Therefore, a) is appropriate. The “sequence” of “sequence control” refers to
controls that progress one after another.

Sequence 1 Sequence 2 Sequence 3

b) This is a description concerning feedforward control. Since the feedback control in c) checks
the result of the controlled variable that is assigned and then performs correction, it must
perform control after the effects of various external sources of interference have appeared.
Feedforward is used to compensate for this flaw in feedback control. However, when only
feedforward control is used, the target value for control cannot be reached in some cases, so
it is normally used together with feedback control.
c) This is a description concerning feedback control. This is a method for control that maintains
a consecutive physical volume that is the subject of control (such as displacement or speed)
at a set level and closes in on a target value.
d) This is a description concerning fuzzy control. This is a type of feedback control and uses the
fuzzy theory to enable control that includes ambiguity. Basically, it is a control method that
attempts control by the representation of not only states such as “much” or “little” but also
intermediate states such as “slightly more” as “more × 0.7” and “slightly less” as “less × 0.7”
with numeric values (probabilities).

Q1-27 a) Operations on stacks

A stack is a data structure that retrieves data with LIFO (Last In First Out). Thus, when data is
retrieved from a stack with a “pop,” the data that was stored in the stack with the “push”
immediately prior is retrieved. The first pop retrieves 2, the second pop retrieves 4, and the third
pop retrieves 5. As a result, 1 and 3 remain in the stack. Therefore, a) is correct. In the text of the
question, the push that disappears with the pop should be confirmed while a diagonal line is
written as below:
push 1→push 2→pop→push 3→push 4→pop→push 5→pop

15
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-28 c) Changes to lists in linked cells

In order to change the list [Tokyo, Shinagawa, Nagoya, Shin-Osaka] to [Tokyo,

Shin-Yokohama, Nagoya, Shin-Osaka], the cell after Tokyo should point to Shin-Yokohama not
Shinagawa, and the cell after Shin-Yokohama should point to Nagoya. The question reveals that
the values in the second column of the table that represents the array represent the number of
linked elements. Data that shows the location of the next item of data is called a pointer.
The operation to make the pointer A(1,2) for A(1,1) = Tokyo point to Shin-Yokohama in row 5
is 5 → A(1, 2), but if this is performed first it results in the disappearance of the value A(1, 2) = 2
that represents the data after “Tokyo.” Due to this, the operation that makes the pointer A(5, 2) for
A(5, 1) = “Shin-Yokohama” point to the next “Nagoya” is performed first. The value (= 3) of the
pointer A(2, 2) for A(2, 1) = “Shinagawa” should be assigned to A(5, 2). The operation in the
options that has the same meaning as A(2, 2) → A(5, 2) is A(A(1, 2), 2) → A(5, 2), so c) is correct.
It should be noted that if the same operations are performed in the sequence in a), 5 is assigned
to A(1, 2) in the first operation, and in the second operation A(A(1, 2), 2) → A(5, 2) becomes A(5,
2) → A(5, 2) so the correct process cannot be performed.

Q1-29 c) Insertion of data into a binary search tree

The conditions for the binary search tree are as shown below:
Value of left child < value of an element of a given node < value of right child
If a tree with the structure of these conditions is selected from the options, the answer is c).
8

3 10

1 5 9 15

13

12

a) The value of the node of the left child of node 10 is 9, and the value of the node of its child is
12. Therefore, the conditions are not fulfilled.
b) The value of the node of the left child of node 10 is 12, and the conditions are not fulfilled.
d) The value of the node of the right child of node 15 is 12, and the conditions are not fulfilled.

16
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-30 c) Hash functions

The key is 1094, so the hash value h(1094) = mod(1094, 97) = 27. In the range of 1through
1000, the keys where the hash value is equal to 1094 are keys where the remainder after division
by 97 is 27. The values that are applicable are multiples of 97 plus 27, and the value 27 itself. As
such, the number of multiples of 97 that are included in the range of 1 through 1000 is first
calculated. 1000 ÷ 97 = 10 (remainder 30), and the number of multiples of 97 is 10. Since key x is
a multiple of 97 plus 27, “ 97 × 10 + 27 ”gives 997, and this is within 1000. Thus, there are 10
values that are a multiple of 97 plus 27, so including 27 the number of keys where the hash value
is equal to 1094 is 11. Therefore, c) is correct.

Q1-31 a) Explanation of the hash method

Hashing is a method for performing a calculation (hash function) on the key value of a record,
and setting the storage location. When data is searched for, by performing the same calculation on
the search key, the storage location can be directly accessed. Therefore, a) is appropriate. Also, in
this method, with a hash function, the storage location of multiple items of data is sometimes the
same. This is called a synonym (collision).
b) This is an explanation concerning a method for dealing with synonyms in hashing called
“chaining”. With chaining, in the storage location calculated with a hash function, a pointer
to synonym data is stored. Also, when a synonym occurs, the stored address is accessed one
after another and the synonym address is stored in the final item of data.
c), d) To calculate an address from the key value of a record, there are indirect address methods
and direct address methods. An indirect method performs a calculation on a key value and
converts it to an address. Hashing is an indirect address method. Direct address methods
include a method that creates a table with corresponding key values and addresses which is
referenced during access (c), and a method that converts a key value into an address (d).

Q1-32 a) Number of comparisons in a binary search

A binary search is a search method where the subject of the search is sorted in advance and then
the range of the search is reduced by repeatedly dividing it in two and searching for the required
data. When the required data certainly exists in n data elements, the range of the search is
repeatedly divided in half and when only one data element remains the required data is found.
Here, when the range of the search is divided in half rounding is performed, so the data can be
considered to be found when the number of data elements in the range of the search is one or more
and less than two.
At this time, if the number of comparisons is k, then 1 ≤ n/2 < 2. If each term is multiplied by
k

2 , it becomes 2 ≤ n < 2×2 (or 2 ≤ n < 2 ＋ ), and by using a logarithm with a base of 2 this can
k k k k k 1

be represented as k ≤ log2n < k+1. The average number of comparisons is an integer value that
fulfills this expression, and is approximately log2n. Therefore, a) is correct.

17
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-33 c) Search processing with binary search

This is a flowchart for a binary search. In a binary search, the data to be searched for is already
sorted. As such, in comparison the data to be searched and the value located at the center of the
search range, it can be understood whether the first half or latter half of the range contains the
search data.
In the question, the search data is x, and the target data is stored in array AR of size n. In the
evaluation immediately before A and B, a comparison of AR(k) with x is made. Here, AR(k) is the
data in the central position of the search range from the previous process [(lo + hi)/2 → k] (lo =
lowest, hi = highest). If the result of comparison is “A(k) < x”, then x is in the latter half. Also, if
the result is “A(k) > x”, x is in the first half. As such, in each of blanks A and B the new search
range is set. In blank A, since the latter half is the new range, the upper limit should be left the
same and the lower limit should be “k + 1” that represents the next value from the central point k.
In blank B, since the first half is the new range, the lower limit should be left the same and the
upper limit should be “k – 1” that represents the value before the central point k. Therefore, c) is
correct.

Q1-34 d) Methods for sorting data

For sorting methods, as well as the three basic sorting methods of bubble sort, selection sort,
and insertion sort, there are applied sorting methods with a reduced volume of calculations
including quick sort, shell sort, and heap sort. Many of the applied sorting methods use recursive
algorithms, and it is necessary to understand the basic flow of each algorithm. The heap sort is an
improvement of the selection sort. The selection sort is a method that takes the smallest (or largest)
data element from a string of data and makes it the first element, and then from the remaining data
string takes the smallest (or largest) data element. A heap sort is a sorting method that is based on
this selection sort and uses a tree structure called a “heap.” Therefore, d) is appropriate.
a) A quick sort is a method that is an improvement of the bubble sort. It is a method that sorts
by repeatedly retrieving a single element of data for comparison and then dividing the data
into a group with a value lower than that element of data and a group with a value less than
that element of data. The description is of a shell sort.
b) A shell sort is a method that is an improvement of the insertion sort. It is a method that
retrieves data in a data string at a certain interval (gap) and makes it a subset. It then sorts
each data string with an insertion sort, reduces the interval further, and sorts the data again.
The description is of a bubble sort.
c) A bubble sort (exchange sort) is a method that compares adjacent elements of data and swaps
these elements of data if the order of size is the wrong way around. The description is of a
quick sort.

18
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-35 c) The data structure used in function and procedure calls

When a function or a procedure is called, the process that is running is temporarily halted, the
data that is being processed is saved, and the new process is started. A process that is called later is
processed first. After this process is finished, the previous process is restarted based on the saved
data of the halted process. Saved data in this kind of situation must be processed with the last in
first out method. A stack has a last in first out data structure. In a stack, the most recent data is
always prioritized over older data and retrieved. Therefore, c) is correct.
a) A binary search tree is a tree structure where the paths from the root to the leaves proceed so
that a node with a smaller key value than the node is linked to one side and a node with a
larger key value is linked to the opposite side. Consequently, a high-speed data search can be
performed using a key value.
b) A queue is a data structure that is waiting to be processed that occurs in transaction
processing. This data structure is processed in a first in first out manner.
d) A doubly-linked list is a data structure where each element of a series of data has two
pointers that can allow forward and backward movement along the elements.

Q1-36 c) Recursively defined functions

The value of f (5) is calculated with a recursive function, but f (5) means n = 5. Given the
expression in the text of the question, when n = 1, 1 is returned as the value, and in other
situations, the value of n + f ( n − 1) is returned. Thus, the equations hold as shown below.
When n=1; f (1) = 1
When n=2; f ( 2) = 2 + f ( 2) = 2 + 1 = 3
When n=3; f (3) = 3 + f ( 2) = 3 + 3 = 6
When n=4; f ( 4) = 4 + f (3) = 4 + 6 = 10
When n=5; f (5) = 5 + f ( 4) = 5 + 10 = 15
As a result, when n=5, f (5) = 15 . Therefore, c) is correct.

Q1-37 d) Relationship of initial values in a flowchart

In order to execute (1) → (2) → (3) → (5), the relationship “m < n” (i.e., “a < b”) must be valid.
Thus, b) or d) is applicable. In (5), “n – m” is assigned to n, and in order to execute (2) → (3) →
(4), “m > n” must be valid at this level. Under these conditions, if the relationship is “2a = b” as in
b), then “m = n” holds in the operation in (5) and execution in the order specified is not possible.
On the other hand, with “3a = 2b” as in d), “a → m” and “1.5a → n” are executed in (1), “n =
0.5a” and “m = a” hold in the first operation of (5), and in the next operation (4) after (2) → (3)
both m and n are 0.5a. As a result, in the evaluation of the third (2), “m = n” can hold and (6) can
be executed. Therefore, d) is correct.

19
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-38 c) Calculation of the total amount of awards with a decision table

Concerning improvement proposals A and B, if both are applied to the conditions of the decision
table, each proposal is as shown below:
• Improvement proposal A
Reduced amount of 200,000 yen → Reduced amount of less than 100,000 yen: No
Reduced period of 3 days → Reduced period of less than one week: Yes
• Improvement proposal B
Reduced amount of 50,000 yen → Reduced amount of less than 100,000 yen: Yes
Reduced period of 2 weeks → Reduced amount of less than one week: No
Thus, the award for each of improvement proposals A and B is 1,000 yen, and the total amount
is d) 2,000 yen.

Q1-39 b) Standardization of programming

The standardization of programming aims for an improvement in the quality of programming. It

sets rules for the writing of a program, and programs are created based on these rules. As such,
anyone who is involved with development and maintenance can easily understand the details of a
program. Also, concerning programming where mistakes are easy to make and programming
where there is a danger of the occurrence of a security problem, the occurrence of such problems
can be prevented by the clarification of the writing method and sharing information. Therefore, b)
is appropriate.
a) Optimization of compilation is not related to the standardization of programming.
c) The standardization of programming generally depends on the programming language.
d) Through the standardization of programming, the promotion of program creation with
efficient processing is possible, but the definition of a standard execution time is not
applicable.

Q1-40 b) Program structure

A reentrant program can be executed by multiple processes simultaneously. In order to enable

this, a program is divided into a procedure part and data part as in the description of b) and a data
part must be held for each process. Therefore, b) is appropriate.
a) A recursive process must be controlled by recording the status of a running program with
LIFO (Last-In First-Out).
c) A serially reusable program can reuse a program that is loaded into main memory without
reloading it, and is not a reentrant program. However, a reentrant program is serially
reusable.
d) This is a description of a reentrant program.

20
 Morning Exam Section 1 Basic Theory Answers and Explanations

Q1-41 c) Characteristics of object orientated programs

The correct description concerning object oriented programs is c). An object is a bundle of data
and methods. A group of such objects is a program.
a) This is a description concerning a data flow program.
b) This is a description concerning a procedural program.
d) This is a description concerning a functional program.

Q1-42 a) Explanation of a Java servlet

A Java servlet is a program that is developed with the Java language and executed on a Web
server. Therefore, a) is appropriate. Servlet is a composite word formed from “server” and “-let”
(meaning “small”) and means “small server.” It is executed on a Web server at the request of a
client.
b) A Java program that is downloaded from a server and executed on the client is a Java applet.
The name “applet,” as with “servlet,” is a composite word formed from “application” and
“-let” and means a small application.
c) Rules for handling a program developed with Java as an application component are called
JavaBeans. JavaBeans is a specific mechanism for component oriented development.
d) One of the characteristics of the Java language is that a module generated by compiling a
source program is an intermediate code called byte code rather than a program in machine
language program that depends on the CPU of the computer it runs on. Then, this
intermediate code is converted by an execution environment (interpreter) called a Java VM
(Virtual Machine) on the computer into machine language program appropriate for the CPU
on that computer, and then executed. This mechanism achieves a feature called
“multiplatform” that does not select the type of computer that a program runs on. In order to
execute a Java servlet, a Java VM is installed on a Web server.

Q1-43 b) Characteristics of XML

HTML (HyperText Markup Language) is a language for the description of webpages. “Tags”
enclosed in angle brackets are used to define the layout of a document. Also, by specifying a URL
(Uniform Resource Locator), a document can “link” to another document.
In initial development, XML (eXtensible Markup Language) was positioned as a meta-language
for the exchange of structured documents on the Web, but it has progressed to a format also for the
exchange of a wide range of data. In addition to the two characteristics (tags and links) of HTML,
the meaning of tags can be defined with the DTD (Document Type Definition) function, and the
structure of the document itself as well as the data can also be defined.
For example, the tag <product info> ... </product info> can be defined and data can be
represented. Therefore, b) is the most appropriate.

21
 Morning Exam Section 1 Basic Theory Answers and Explanations

a) XML complements the weakness of HTML, which is not suitable for the representation of
data or document structure, and does not aim for improvement in the performance of
webpage display.
b) The style sheet languages that can specify the style of an XML document include XSL
(eXtensible Style Language) and CSS (Cascading Style Sheets). In order to display an XML
document on a Web browser, these convert XML to HTML.
d) Like HTML, XML was developed based on SGML (Standard Generalized Markup Language).
For example, the DTD function of XML is a function that SGML has.

22
 Morning Exam Section 2 Computer System Answers and Explanations

21Section 2 Computer System

Q2-1 a) Execution order of an instruction involving access to the main memory

An instruction is executed in the order of “instruction fetch →instruction decode → effective

address calculation →operand fetch → execution.” Therefore, the “operation that is performed
between “instruction decode” and “operand fetch” is a) Calculation of an effective address. Here,
an operand refers to the data on which the instruction is to be processed.
This operand uses various types of addressing modes, which are classified as follows:
• Direct addressing: specification of an address in the main memory
• Index addressing: specification of an address in the main memory by using the result of
adding the value of an index register to the value of the address part
• Relative addressing: specification of a position relative to the position where the its own
instruction is stored
• Indirect addressing: specification of the address of the target data in the main memory by
using a value stored in an address in the main memory that is indicated in the address part
• Register addressing: specification of an address stored in a register
• Immediate addressing: specification of the value of the address part as the target data

Q2-2 c) Cause of an external interrupt

Interrupts include internal interrupts that occur as a result of the execution of an instruction
within a program, and external interrupts that occur regardless of an executing instruction. Among
the options, an interrupt that occurs when a hardware failure is detected as described in c) is
classified as an external interrupt. Each of the interrupts mentioned above includes the following:
• External interrupts: An I/O interrupt (e.g., completion of an I/O operation or malfunction of
the I/O device), a timer interrupt (e.g., timeout of a timer), and a machine check interrupt
(e.g., hardware malfunction or power failure)
• Internal interrupts: A supervisor call interrupt (e.g., usage of an OS function by a SVC
instruction or system call instruction of the running program), and a program interrupt (e.g.,
divide-by-zero, overflow, execution of an invalid instruction, or page fault)
a) A page fault that occurs when a page that does not exist in the main memory is accessed by a
program instruction is classified as an internal interrupt.
b) Generally, a system management instruction, which is an OS function, is executed in kernel
mode (also called the privileged mode and supervisor mode), and if a privileged instruction
such as a system management instruction is executed under the user mode in which general
application programs are executed, a privileged instruction violation occurs. Since this
privileged instruction violation is a program interrupt, it is classified as an internal interrupt.
d) A floating-point operation is performed by a program instruction, and as a result, an overflow
generates a program interrupt classified as an internal interrupt.

23
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-3 a) Description concerning the CPU clock frequency of a PC

Clock frequency is the frequency of a signal (clock) generated to determine the timing of an
internal operation of a computer, and is expressed in MHz (megahertz) and GHz (gigahertz). One
MHz indicates that the signal is generated one million times in one second, and one GHz means
that the signal is generated one billion times in one second.
Since the CPU operates synchronously with the clock, generally, the speed of execution of an
instruction increases with the frequency. Therefore, the description of a) is appropriate.
b) The clock frequency is related to the CPU operation, and does not control the communication
speed.
c) The clock frequency is related to the CPU operation, and the rotation speed of a disk is a
value specific to a device.
d) The clock frequency is related to the CPU operation, and the interrupt interval of real-time
processing is not directly related to the clock frequency.

Q2-4 d) Calculating the number of instructions that can be executed in one second

A clock is a signal for determining the timing of an internal operation of a CPU. The clock
frequency represents the number of signals generated in one second in MHz (megahertz) or GHz
(gigahertz). Normally, a single instruction consists of several basic operations, which are
performed in one clock cycle, and therefore, one instruction is executed in several clock cycles.
However, when multiple instructions of machine language are executed simultaneously by using a
technique called superscalar, the average number of clock cycles for the execution of an
instruction can be reduced to below one.
Since the CPU runs at one GHz, this CPU generates one giga clock signals in one second. One clock
−9
period is 1 (second) ÷ 1 G = 10 (seconds). Next, since this CPU “is capable of executing one
machine instruction in 0.8 clock cycles on average,” the average execution time of one instruction
-9
is 0.8×10 seconds. Therefore, the number of instructions that can be executed in one second is as
follows:
−9
1 (second) ÷ (0.8×10 ) (seconds/instruction)
= 109 ÷ 0.8 = 1,000,000,000 ÷ 0.8
= 1,250 million instructions)
Therefore, d) is the correct answer.

24
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-5 c) CPU performance in MIPS

9
A clock frequency of 1 GHz indicates that clock signals are generated 1 × 10 times in one
second. The average instruction execution time (average number of clock cycles necessary for
executing one instruction) of the CPU is calculated from the table of assigned instruction types,
and the performance is calculated by dividing the clock frequency by the average instruction
execution time.

Average instruction execution time

= 10 × 0.6＋5 × 0.4
= 8 (clocks)
9
Performance = (1 × 10 ) ÷ 8
= 125,000,000 (instructions/second)

This becomes 125 MIPS when it is converted to MIPS (Million Instructions Per Second).
Therefore, c) is the correct answer.

Q2-6 c) Pipeline processing in a processor

As shown in the figure below, pipeline processing is a method in which a single processor
executes multiple instructions simultaneously by slightly shifting them in stages. Therefore, c) is
the correct answer.

IF ID OA OF EX RS IF : Instruction Fetch
IF ID OA OF EX RS ID; Instruction Decode
IF ID OA OF EX RS OA: Operand Address Calculation
IF ID OA OF EX RS OF: Operand Fetch
IF ID OA OF EX RS EX: Execution
IF ID OA OF EX RS RS: Storing operation results
A maximum of six in parallel

Example of pipeline processing

a) This is an explanation of the SIMD (Single Instruction stream Multiple Data streams)
method. This method is suitable for matrix operations.
b) This is an explanation concerning a processor in which pipeline processing is not adopted.
d) This is an explanation of the MIMD (Multiple Instruction streams Multiple Data streams)
method. This method is applicable to a multiprocessor system.

25
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-7 c) Explanation of superscalar

Superscalar is the technique of simultaneously executing multiple instructions by further

dividing the pipeline mechanism into multiple stages with increased processing performance
during the execution of multiple instructions. Therefore, c) is the correct answer. In the CPU, when
multiple instructions are to be processed in parallel, the simultaneous execution of interrelated
multiple instructions must be avoided. In superscalar, the CPU performs this control in the
hardware. VLIW (Very Long Instruction Word), on the other hand, omits the time-consuming
operation of determining the possibility of parallel processing by the CPU through the compilation
of an extremely large group of instructions that must be processed in sequence into a single
instruction.
a) This is an explanation concerning a vector processor.
b) This is an explanation concerning a super-pipeline.
d) This is an explanation concerning VLIW.

Q2-8 c) Types of CPU

The method that allows one instruction to perform multiple instances of the same operation
simultaneously for different data is c) SIMD (Single Instruction stream Multiple Data streams). SIMD
is suitable for multimedia data processing in which the same operation is often repeated for continuous
data, and typical examples include processors called GPU (Graphics Processing Unit) that are used
exclusively for image processing and vector computers.
All options are a classification of CPU called the Flynn’s taxonomy, where S indicates Single,
M indicates Multiple, I indicates Instruction, and D indicates Data, and if it is understood that the
options are a combination of these, all options need not be memorized. The details other than the
correct answer are as follows:
a) MIMD (Multiple Instruction streams Multiple Data streams)
Multiple processors process multiple data in parallel based on individual instructions for
each data. This method is applicable to a general multiprocessor.
b) MISD (Multiple Instruction streams Single Data stream)
A single piece of data is processed by multiple instructions, but a processor based on this
method is not yet implemented in a practical way.
d) SISD (Single Instruction stream Single Data stream)
A single piece of data is processed by a single instruction. This method is applicable in a
general computer that is controlled sequentially by a single processor.

26
 Morning Exam Section 2 Computer System Answers and Explanations

SIMD method MIMD method

Data A Instruction 1 Data A

Instruction Data B Instruction 2 Data B

Data C Instruction 3 Data C

Processor Processor
Different processors execute the same Different processors execute different
instruction for different data. instructions for different data.

Q2-9 a) Explanation of DRAM

As described in a), DRAM (Dynamic Random Access Memory) is memory that stores one (1) bit
depending on whether or not each capacitor is electrically charged. Since DRAM is composed of
tiny capacitors, the stored charge is discharged over time, and an operation called “refresh” by which
the content is rewritten every few milliseconds is necessary. Although the access speed is slower
than SRAM (Static RAM) because of this refresh operation, a simple circuit makes the unit cost per
bit low, and memory with a large capacity is easy to manufacture. Therefore, main memory is
usually composed of DRAM chips.
b) It is an explanation of mask ROM (Read Only Memory).
c) It is an explanation of EPROM (Erasable Programmable ROM).
d) It is an explanation of SRAM.

Q2-10 b) Storage device with the shortest access time

In the storage hierarchy, the closer a storage device is to the CPU, the shorter the access time is
in general, and the one-time access amount is also small. Inside the CPU, the register that is
always used when instructions are executed has the shortest access time, and therefore, b) is the
correct answer.
When storage devices are arranged in order of shortest access time first after the register, the
order is generally as described below. The approximate access time of each device and the storage
capacity must be known.

CPU register < primary (L1) or secondary (L2) cache memory of CPU
< main memory < hard disk < (optical disk such as CD and DVD) < (magnetic tape)

27
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-11 a) Description concerning the cache memory

The cache memory is a high-speed buffer storage (a holding area for temporary storage) that is
implemented for bridging the difference in the speed of the main memory and CPU. By deploying
the memory content that has a high probability of being accessed in cache memory, the access
frequency to the main memory is reduced and speed is improved. When a write instruction is
executed, the “write through” method rewrites both the cache memory and main memory, and the
“write back” method rewrites the cache memory first, and then the main memory when the
relevant data is flushed from the cache memory. Therefore, a) is appropriate.
b) When a cache miss occurs, generally the data is processed in the hardware within the
instruction execution cycle, and an interrupt does not occur.
c) The purpose of the cache memory is to bridge the difference in the speed of the main memory
and CPU. It is not meant to bridge the difference in the capacity of the real memory and
virtual memory. In reality, the capacity of the cache memory is very small as compared to the
capacity of the real memory.
d) Along with an improvement in the access speed of the semiconductor memory, the operation
speed of the CPU is also improving. Because of this, the difference in the speed is not being
bridged. As a result, the necessity of the cache memory is not decreasing.

Q2-12 d) Combination having the shortest effective access time

In order to bridge the difference in the processing performance of the CPU and the access speed
of main memory, a cache memory that supports high-speed access is placed between the two. The
hit ratio is the probability of existence of the target data in the cache memory. If the target data
does not exist in the cache memory, the data is read from the main memory, but if the hit ratio is p,
the probability will become 1 − p.
If the access time of the cache memory is Tc and the access time of the main memory is Tm, the
effective access time is Tc × p + Tm × (1 − p). If calculation is performed based on the values of
each option, d) has the shortest access time as shown below:
a) 10 × 0.6 + 70 × (1 − 0.6) = 6 + 28 = 34 (nanoseconds)
b) 10 × 0.7 + 70 × (1 − 0.7) = 7 + 21 = 28 (nanoseconds)
c) 20 × 0.7 + 50 × (1 − 0.7) = 14 + 15 = 29 (nanoseconds)
d) 20 × 0.8 + 50 × (1 − 0.8) = 16 + 10 = 26 (nanoseconds)

28
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-13 d) Calculating the hit ratio of the cache memory

If the access time of the cache memory is Tc, the access time of the main memory is Tm, and the
hit ratio is h, the effective access time is Tc × h + Tm × (1－h). Since the hit ratio and effective
access time of both systems are the same, h can be calculated as follows:
Effective access time of system A = Effective access time of system B
15 × h + 50 × (1 − h) = 10 × h + 70 × (1 − h)
15h + 50 − 50h = 10h + 70 − 70h
25h = 20
h = 0.8.
Therefore, the hit ratio is d) 0.8.

Q2-14 b) Explanation of memory interleaving

Memory interleaving is a technique for the improvement of speed of access to main memory.
Main memory is internally divided into multiple sections called banks that operate independently,
and as shown in the figure below, by arranging adjacent addresses in separate banks, high speed is
achieved through concurrent access. Therefore, b) is appropriate. Memory interleaving is a
technique that has been used in general-purpose computers, but currently, it is used in some
servers and PCs as well.

Processing unit (CPU)

Main memory Address 0 Address 1 Address 2 Address 3

Address 4 Address 5 Address 6 Address 7
: : : :
Bank 1 Bank 2 Bank 3 Bank 4

a) When the data to written to the cache memory by the CPU, it is also written to main memory.
This method is called “write through.” According to this method, the content of both the
cache memory and main memory always match, but because data is also written to the main
memory every time it is written to the cache memory, the processing speed during writing
declines.
c) It is a description concerning the purpose of the cache memory.
d) In the case of CPU architecture in which an instruction cannot be rewritten, if the instruction
cache and data cache are separated, writing to the instruction cache is not necessary, which
leads to an overall improvement in the cache efficiency. However, this is not directly
connected to the method for removing the cause of pipeline processing disruption (such as
inclusion of several branch instructions), and is therefore not related to memory interleaving.

29
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-15 a) Optical disc using organic dye and laser beam

An optical disc that uses organic dye as the recording layer of the storage media, and records
data with a laser beam by burning tiny holes called pits is a) CD-R (CD-Recordable). CD-R is a
type of CD that allows data to be written only once, and because of burning pits, once data is
written, it cannot be erased.
b) CD-RW (CD-ReWritable) is a type of CD on which data can be written and erased any
number of times. Data is recorded by increasing the temperature with a laser beam, and then
changing the properties (degree of reflection) of the material of the recording layer.
c), d) DVD is a storage medium that has the same 12 cm diameter as CD. The recording density
is higher in comparison with CD, and depending on the type of DVD, recording on both sides
or dual-layer recording is possible. DVD-RAM described in c) is a type of re-writable DVD,
and DVD-ROM described in d) is a type of read-only DVD. The method of recording data on
DVD-RAM is almost same as CD-RW, and on DVD-ROM, data is recorded by making tiny
dents called pits in the same way as in CD-ROM.

Q2-16 c) Explanation of a system bus

A system bus is a common path (bus) used by the CPU to send data and control signals between
main memory and an I/O controller. Therefore, c) is appropriate. Buses are classified with various
methods, but if the buses are divided into a computer’s internal buses and external buses, the
system bus is an internal bus that is the same as the processor bus (connection between the
components of the CPU) and memory bus (connection between the CPU and main memory). The
external buses include I/O buses for connections between an I/O controller and auxiliary storage,
and also for connections between I/O controllers.
a) This is a description concerning RS-232C.
b) This is a description concerning DMA (Direct Memory Access) and I/O channels.
d) This is a description concerning USB. The high-speed mode (480 Mbps) was added from
USB2.0, and the super-speed mode (5 Gbps) was added from USB3.0.

30
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-17 c) Explanation of USB

USB (Universal Serial Bus) is a serial interface that

enables the PC and all peripheral devices to be connected Hub
PC
via the same connector. By using a USB interface, a
maximum of 127 devices can be connected in a tree Hub
topology via a hub (this is called the cascade
connection).Therefore, c) is the correct answer.
Three types of transfer speeds, namely the low speed (1.5 Mbps), full speed (12 Mbps), and high
speed (480 Mbps) are available, and the high-speed mode is supported only by USB2.0. The hot
plug is supported, which enables removal and insertion of the connector while power is being
supplied to the computer. The super-speed mode (5 Gbps) has been added to USB3.0, with a
power supply increased from 500 mA to 900 mA.
a) The parallel interface used for connecting the PC and the CD-ROM or DVD drive provided
in the PC is ATA/ATAPI (AT Attachment/AT Attachment Packet Interface).
b) The parallel interface used for connecting the hard disk or a printer through a daisy chain is
SCSI (Small Computer System Interface; pronounced “skuzzy”) defined by ANSI (American
National Standards Institute).
d) The serial interface that transfers data to a printer by using infrared rays is IrDA (Infrared
Data Association).

Q2-18 d) Characteristic of serial ATA

The IDE (Integrated Drive Electronics) specification, which is the industry common standard
for hard disk connection interfaces in PCs, standardized by ANSI (American National Standards
Institute) is referred to as the (parallel) ATA (Advanced Technology Attachment) standard.
According to this standard, up to two built-in hard disks can be connected within a computer
through parallel transfer. Following this, extended standards such as E-IDE (Enhanced IDE) and
ATAPI (ATA Packet Interface) were defined, which have enabled the connection of CD and DVD
drives besides hard disks.
Serial ATA is aimed at a reduction in operating voltage, reduction in size of the connector and
cable, and increase in the transfer speed by using simple serial transfer as the transfer method, and
has a transfer speed of 1.5 Gbps and 3 Gbps. Furthermore, the hot swap function that enables
removal and insertion of connectors while power is supplied was not supported by parallel ATA,
but is supported by serial ATA. Therefore, d) is appropriate.
a) The SAS controller has upward compatibility with serial ATA, but does not have mutual
compatibility.
b) This refers to SCSI.
c) In comparison with parallel ATA, the cable thickness and the connector shape are very
different.

31
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-19 b) Software serving as an interface between the OS and a peripheral device

In peripheral devices such as a printer, the details of control methods are different for each
model, which makes it impossible to include the control functions of all models beforehand in the
OS, and also leads to inefficiency. Therefore, rather than including the control program of the
peripheral devices in the OS, a program is made available for each peripheral device as necessary.
Such a program that acts as an interface between the OS and a peripheral device is called a “device
driver,” so b) is the correct answer.
a) This is a program that enables software to be embedded in the hard disk for use, and allows
the copying of necessary files and specification of the environment settings.
c) This is a function that is embedded in the OS that manages the connection state of peripheral
devices, settings of the drivers, and settings of the interrupt request signal (IRQ) from the
peripheral devices.
d) This is a program that initializes (or formats) media such as floppy disks, hard disks, and MO
disks to enable them to be used.

Q2-20 d) Display characterized by a low-voltage drive and low power consumption, which does not require

The display that emits light by itself when electric voltage is applied, and satisfies all
characteristics including “no backlight required,” “low voltage operation,” and “low power
consumption” is an Organic Light Emitting Diode (Electro Luminescence), therefore, d) is the
correct answer. The OLED display has a structure in which an organic compound that emits light
in response to an electric voltage is sandwiched between glass (or plastic) substrates, and displays
images when a 5 to 10 volt of direct-current voltage is applied. OLED displays are also available
as thin types that have a thickness of 1.8 mm. Unlike liquid crystal displays, a backlight is not
needed, which leads to a low power consumption.
a) CRT (Cathode Ray Tube): This refers to the brown tube in TVs. A surface is coated with a
fluorescent material, and an electron beam is deflected by a magnetic field or electrode and
made to hit the surface so that the fluorescent material glows. The power consumption is
higher than liquid crystal.
b) PDP (Plasma Display Panel): This is a display in which a gap between two glass plates is
filled with gas such as helium or neon, and light is emitted by applying an electrical voltage
to the gas. The display is thin, lightweight, and has a high brightness, but consumes more
power than liquid crystal.
c) TFT liquid crystal: This is a display device in which a substance called liquid crystal, whose
molecular structure changes in response to electric voltage and the degree of light
transmission varies, is inserted between glass substrates. A liquid crystal in which electric
voltage is applied by a thin film transistor to each bit of a pixel is called TFT (Thin Film
Transistor). Since liquid crystal does not emit light by itself, it needs a backlight.

32
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-21 a) Explanation of plasma display

A plasma display is a display in which a gap between two thin glass plates is filled with neon
gas or xenon gas, and an electrical discharge is produced by applying a high voltage to the gas.
Ultra violet rays that are generated by this electrical discharge then hit the red, green, and blue
(RGB; three primary colors of light) fluorescent material on the surface of the glass and light is
emitted. Therefore, the description in a) is appropriate. At present, the power consumption is not
high, and in some cases the power consumption is actually reduced by the addition of the backlight
used in liquid crystal displays.
b) This is a description concerning liquid crystal displays such as a TFT (Thin Film Transistor).
Since the display itself does not emit light, a backlight is installed at the back or side of the
display as a light source.
c) This is a description concerning an OLED (Electro-Luminescence) display. An OLED has a
structure in which a thin film of an organic substance is sandwiched between electrodes, and
this structure has been put into practical use in some places. Unlike a liquid crystal display
such as a TFT, a light source is not needed, so the structure can be slimmed down so far as to
become a film.
d) This is a description concerning the brown tube (CRT; Cathode Ray Tube).

Q2-22 b) Access time of a hard disk

The average access time of a hard disk can be calculated with the following formula:

Average access time = Average seek time + Average rotational latency + Data transfer time

The average seek time is 20 milliseconds.

The average rotational latency is the time required for half (1/2) rotation.
The rotation speed of the disk is 5,000 rpm, that is, 5,000 rotations in one minute (60 seconds).
Time required for one rotation = 60 ÷ 5,000 (seconds)
= 0.012 seconds
= 12 milliseconds
Average rotational latency = 12 milliseconds ÷ 2 = 6 milliseconds
If the data transfer time is calculated before the data transfer time is calculated, 15,000 bytes of
storage capacity of one track is transferred in 12 milliseconds, which is the time for one rotation,
and therefore, the data transfer speed per millisecond will be as follows:
15,000 ÷ 12 (bytes/millisecond) = 1,250 (bytes/millisecond).
And since one block is 4,000 bytes, the data transfer time will be as follows:
Data transfer time = 4,000 (bytes) ÷ 1,250 (bytes/milliseconds)
= 3.2 (milliseconds).
Based on the above, the average access time is
20 + 6 + 3.2 = 29.2 milliseconds
Therefore, b) is the correct answer.

33
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-23 d) Calculating the number of sectors

Since one block is composed of 8 sectors, the memory capacity per block is
500 bytes × 8 sectors/block = 4,000 bytes/block.
Since file area is allocated in units of blocks, 2,000 bytes ÷ 4,000 bytes/block = 0.5 (rounded up
to 1), to indicate that 1 block is used to store a file of 2,000 bytes. Also, 9,000 bytes ÷ 4,000
bytes/block = 2.25, which is rounded up to 3, to indicate that 3 blocks are necessary to store a file
of 9,000 bytes.
Therefore, a total of 4 blocks are allocated, and the necessary number of sectors is 4 blocks × 8
sectors/block = 32 sectors. Therefore, d) is the correct answer.

Q2-24 a) Method of segmenting and storing data on multiple hard disks

Striping refers to the distribution and recording of data of a constant size and the recording it on
multiple hard disks so that data is read and written in parallel. In RAID (Redundant Array of
Inexpensive Disks), striping corresponds to RAID0. Therefore, a) is the correct answer.
b) Disk cache: This is a device or function that increases the speed of data exchange between a
CPU and disk. The data that is transferred as a block from the hard disk can be stored in the
cache to increase the speed of reading.
c) Blocking: The logical unit of data for I/O is called a block, and collecting several records as
blocks is called blocking. For sequential access, processing efficiency can be improved in
comparison with cases where data is read record-by-record.
d) Mirroring: This is a method by which exactly the same data is written to multiple hard disks,
and even if a failure occurs in one disk, the data processing is not affected. Mirroring
corresponds to RAID1.

34
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-25 c) Differentiation of RAID configuration

RAID (Redundant Array of Inexpensive Disks) is a technology for improving the speed and
reliability of access by consolidating multiple hard disk drives into a single logical disk. The
access speed is improved by distributing and storing data in multiple disks and enabling access in
parallel (striping). Furthermore, reliability is improved through mirroring of the disk and also by
securing data for error correction. Depending on the unit of striping and the method of securing
reliability, RAID is classified into six types, namely RAID0 through RAID5. Therefore, c) is the
correct answer.
· RAID0: This is a method that performs only striping. Therefore, there is no improvement in
reliability.
· RAID1: This is a method that performs only disk mirroring.
· RAID2: This is a method by which striping is performed at the bit level, data for error correction is
secured with a dedicated (fixed) disk, and error correction is performed by the hamming
method, which is currently still not in practical use.
· RAID3: This is a method by which striping is performed at the bit level, data for error correction
is secured with a dedicated (fixed) disk, and error correction is performed by the parity
method. By using the parity method instead of the hamming method, the redundant data
(data for error correction) can be reduced in comparison with RAID2, which makes this a
feasible method. Initially, parity was performed at the bit level, but in practice it is
performed at the byte level.
· RAID4: This is a method by which striping is performed at the block level, data for error
correction is secured with a dedicated (fixed) disk, and error correction is performed with
the parity method, which is currently still not in practical use.
· RAID5: Striping is performed at the block level, and data for error correction is distributed and
deployed on each disk. By the distribution and deployment of data for error correction,
access to the disk in which the data for error correction is stored, the flaw of RAID4 where
access speed is reduced by a bottleneck is improved. RAID5 has been put to practical use
together with RAID1.

35
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-26 c) Configuration of an on-line processing system composed of two systems

A system configuration composed of two systems as shown in the figure in the question is the
most common configuration of an on-line system called a duplex system. When the primary
system that performs on-line processing breaks down, the processing is switched to the backup
system and can be completed by losing only the switching time, which makes this a highly-reliable
system. Therefore, c) is the correct answer.
a) Simplex system: This is a simple system configuration without any redundancy. If any
component breaks down, the entire system breaks down.
b) Dual system: This is a system in which two computers perform the same processing, and the
processing results are collated at a fixed interval. Even if one computer breaks down, the
processing continues with the other computer. This is a system that has a very high reliability with
no loss resulting from a failure, but the cost is very high and the cost performance is low.
d) Parallel processor system: This is a system composed of multiple processors. In the recent
years, multiprocessor workstations with multiple (two to four) processors installed have even
appeared, but a parallel processor system indicates a system with a relatively large number of
processors, such as a super computer.

Q2-27 a) Hot standby system

A hot standby system is one of the techniques for improving reliability. A computer system is
duplicated, the same application software as the primary (or currently used) system is also
installed on the backup system which is left in a standby state, and as soon as a failure occurs in
the primary system, the processing is switched to the backup system and continued. Since both the
primary system and the backup system operate from the time normal operation starts, the
reliability improves, but a certain amount of cost is also involved. On the other hand, a system in
which the backup system does not operate but is left in a standby state is called a cold standby
system. Under these conditions, the cost is low in comparison with a hot standby system, but some
time is taken up in switching the system, so the system stops during the switching time. Therefore,
a) is appropriate.
b), c) In a hot standby system, the backup system operates when the primary system fails (breaks
down).
d) The backup system is in the standby state when the primary system is operating.

36
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-28 a) Characteristic of a client/server system

A client/server system is a distributed processing system in which a dedicated computer (server) that
provides various services and another computer (client) that requests and uses a service operate in
coordination while processing is shared over a network. There are various types of server such as a
Web server, a mail server, and a database server. Therefore, a) is appropriate.
b) A client/server system is a distributed processing system, and generally, the server accesses
the data resources.
c), d) A client is a computer that uses the service.
Q2-29 a) Client/server system using a browser

It is clear from a “client/server system using a browser” that the software for business operations are
consolidated on the server, and the client only displays the screens via the browser.
In a client/server system that does not use a browser, software for business operations is
necessary for each client, and therefore, maintenance tasks such as version upgrade of the software
for business operations must be performed for each client. However, when a browser is used and
the software for business operations is consolidated on the server, maintenance needs to be
performed only on the server, which reduces the maintenance activities on the client side.
Therefore, a) is the correct answer.
b) Recovery of the server is performed even in systems using a browser.
c) Construction of a database is necessary, regardless of whether a browser is used or not.
d) When a browser is used, the maintenance management of login accounts may actually
increase in order to prevent unauthorized use.

Q2-30 c) Technique of improving the access speed of a hard disk

The sequential distribution of data that has been divided into fixed units (such as bits, bytes, or
blocks) on multiple disk drives to enable parallel access is called disk striping as described in c). It
is the fundamental technique for a disk array.
a) “Disk at once” is a method for writing to a CD-R according to which no additions can be
made once data has been written on it, regardless of the amount of written data. In contrast,
the method by which data is written on separate tracks to enable additions to be made is
called “track at once”.
b) Disk cache is a buffer storage placed between the main memory and disk drive, which can be
accessed at a higher speed than the disk. Data that has been accessed is retained in the disk
cache and imported from there to main memory when the next access request is received. At
this time, since the disk is not accessed, the apparent access time of the disk drive can be
shortened. Two methods are available to implement this. In one method, software uses part of
the main memory, and in the other method an external semiconductor memory in the form of
hardware is used.
d) Disk mirroring is a method of duplicating the same data on two disk drives. Even if one disk
fails, the loss of data can be prevented by using the other disk, and processing can be
continued.

37
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-31 a) Explanation of a fault tolerant system

Fault tolerance is the ability to handle failures. A fault tolerant system is set up in consideration
of accomplishment of a business operation even when the system partially fails. Therefore, a) is
appropriate.
b) This is a description concerning a backup site.
c) This is a description concerning a (loosely coupled) multiprocessor system or cluster system.
d) This is a description concerning a dual system.

Q2-32 a) Concept of fail safe

Fail safe is the concept of controlling a system in a predetermined safe state so as to prevent any
risks as a result of a failure in the system. Therefore, the concept of a) is appropriate. This concept
is adopted in systems such as traffic signal control, where a failure of the system poses a threat to
life.
b) This refers to degraded (fall back) operation, and a system that has this function is called a
fail soft system.
c) This is the concept of a redundant system, and is the basic concept of fault tolerance.
d) In a system operated by an unspecified number of users, the entry of unexpected values and
operating methods need to be given consideration. When such an unexpected operation is
performed after a strict input check is conducted, the concept of making the occurrence of
malfunctions unlikely is called a foolproof design.

Q2-33 d) Method of storage composed of a high-speed dedicated network

The network to connect storage devices, such as a hard disk drive or a magnetic tape unit,
separately from the normal LAN is called SAN (Storage Area Network) d). In the majority of the
cases, this storage is configured by connecting one storage device with another through an optical
fiber channel to form a network with a higher speed than the normal LAN.
The meaning of the other terms is as follows:
a) DAFS (Direct Access File System): This is a file sharing protocol focusing on high-speed
data transfer that is applied in NAS described in c) below, which suppresses the CPU load
and overheads by performing direct communication between applications without going
through the OS.
b) DAS (Direct Attached Storage): This is storage such as a disk array that is directly connected
to the server.
c) NAS (Network Attached Storage): This is a product designed exclusively for storage that
enables file transfer only by connecting directly to a network such as LAN.

38
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-34 d) Explanation of throughput

Throughput is the processing amount per unit time, and “the number of jobs processed per unit
time” described in d) corresponds to throughput. Generally, the maximum number of jobs that can
be executed simultaneously is fixed, and therefore, each job is completed in the shortest possible
time to improve throughput. On the other hand, spooling is a mechanism for removing the
operating time of a low-speed peripheral device, such as a printer, from the execution time of a
job, and by using this mechanism, the execution time of the job can be reduced. This is useful in
improving throughput. Therefore, d) is the appropriate description.
a) This is an explanation of the turnaround time. Turnaround time is a system performance
evaluation indicator that is seen mostly in a batch processing system.
b) Such an index is not common. It is in no way an explanation of throughput.
c) The description itself is correct, but it is not an explanation of throughput.

Q2-35 c) Performance evaluation based on a benchmark

A benchmark test is used to compare and evaluate the performance of a computer by measuring
the execution time of standard programs. Therefore, if various types of benchmark tests are
performed, the characteristics of a system can be understood by using the results, which implies
that the description “useful in selecting the model to be installed” provided in c) is appropriate.
The other descriptions contain the errors described below:
a) The results of the TPC benchmark are expressed in terms of performance value and cost
performance, and a cost scale is used.
b) For example, Dhrystone is used to measure a computer’s performance of integer operations,
and is not a benchmark test for evaluating the performance of the entire computer system.
d) A benchmark test is used to evaluate a target with a specific purpose, and cannot be referred
to as a general evaluation model.

Q2-36 b) MTBF and MTTR

MTBF is the mean time between failures (the mean of the time period during which operation
can be performed continuously), which is calculated as the mean of t 1 through t n.

n n
Mean of t 1 through t n = (Sum of t 1 through t n) ÷ n = ∑ ti ÷ n = 1
n ∑ ti
i =1 i =1

Similarly, MTTR is the mean repair time, which is calculated as the mean of r 1 through r n.
n n
Mean of r 1 through r n = (Sum of r 1 through r n) ÷ n = ∑ ri ÷ n = 1n ∑ ri
i =1 i=1

Therefore, b) is the correct answer.

39
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-37 a) Scale representing availability in RAS

RAS, a set of indexes or indicators representing the reliability of a system, is formed from the
first letter of the words Reliability, Availability, and Serviceability.

This expresses how unlikely a failure is. It is expressed by using

Reliability MTBF (Mean Time Between Failures).

This expresses the possibility of use when the system is needed. It is

Availability expressed by using the availability ( MTBF
).
MTBF＋MTTR

This expresses the ease of repair. It is expressed by using MTTR

Serviceability (Mean Time To Repair).

Therefore, a) is the correct answer.

b), c), and d) Even when the total operating time (MTBF+MTTR), MTBF, and MTTR are known,
how long normal operation would continue cannot be determined.

Q2-38 c) About availability

This question focuses on change in availability. As this question deals with either doubling or
halving MTBF (Mean Time Between Failures) and MTTR (Mean Time To Repair), the answer
can be determined quickly by using specific values. It is understood from below that the
availability increases in case of c).
When availability is calculated with an MTBF of 80 and an MTTR of 20,
Availability = MTBF / (MTBF + MTTR) = 80 / (80 + 20) = 80 / 100 = 0.8
a) When both MTBF and MTTR are doubled such that MTBF becomes 160 and MTTR
becomes 40,
Availability = 160 / (160 + 40) = 160 / 200 = 0.8 (does not change)
b) When both MTBF and MTTR are halved such that MTBF becomes 40 and MTTR becomes
10,
Availability = 40 / (40 + 10) = 40 / 50 = 0.8 (does not change)
c) When MTBF is doubled and MTTR is halved such that MTBF becomes 160 and MTTR
becomes 10,
Availability = 160 / (160 + 10) = 160 / 170 = 0.9 (increases)
d) When MTBF is halved and MTTR is doubled such that MTBF becomes 40 and MTTR
becomes 40,
Availability = 40 / (40 + 40) = 40 / 80 = 0.5 (decreases)
For a general understanding, change the expression for calculating the availability as follows:
Availability = MTBF / (MTBF + MTTR) = 1 / (1 + MTTR / MTBF)
According to this expression, when MTBF is increased and MTTR is shortened, the value of (1
+ MTTR / MTBF) in the denominator becomes less, which means that the availability that is
reciprocal to this value increases.

40
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-39 d) Availability of an entire system

The expression for calculating the availability of a system is different when devices are
connected in parallel and when devices are connected in series.
When devices are connected in parallel:
1 − (1 − Availability of device A) × (1 − Availability of device B)
When devices are connected in series:
Availability of device A × Availability of device B
The devices A and C, and B and D are connected in parallel, and these two sections are
connected in series. Each of the sections in which devices are connected in parallel is available
when either of the two devices is operating, that is, this section operates as long as both devices
have not broken down. In consideration of the conditions of the question, the availability of each
of the parallel sections is expressed as follows:
Section in which A and C are connected in parallel:
1 − (1 − 0.9) × (1 − 0.9) = 1 − 0.01 = 0.99
Section in which B and D are connected in parallel:
1 − (1 − 0.8) × (1 − 0.8) = 1 − 0.04 = 0.96
Since these two sections are connected in series, the availability of the entire system is:
0.99 × 0.96 = 0.9504
Therefore, the closest is d) 0.95.

Q2-40 d) Calculating the printer availability

When the operational state of two printers is expressed by ○ for an available state and × for an
unavailable state, the following four cases are assumed:
(1) Printer with an availability of 0.7 ○
Printer with an availability of 0.6 ○
(2) Printer with an availability of 0.7 ○
Printer with an availability of 0.6 ×
(3) Printer with an availability of 0.7 ×
Printer with an availability of 0.6 ○
(4) Printer with an availability of 0.7 ×
Printer with an availability of 0.6 ×
The cases in which either of the two printers is available and the other is unavailable are cases
(2) and (3). Since the probability of a failure is (1 − availability), the probability of each of the
above cases is calculated as below:
(2) 0.7 × (1 − 0.6) = 0.28
(3) (1 − 0.7) × 0.6 = 0.18
The probability of (2) or (3) is calculated by adding the two cases, which is 0.28 + 0.18 = 0.46.
Therefore, d) is the correct answer.

41
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-41 d) Calculating the average processing time of a transaction

According to the conditions described in the question, the processing time of this transaction is
the sum total of the time of execution of 600,000 instructions and the time required for two file
accesses.
First, since the average access time of the disk is 30 milliseconds, the time required for two file
accesses is calculated as 30 × 2 = 60 milliseconds.
Next, since the CPU performance is 30 MIPS, that is, 30 million instructions can be executed in
one second, the time required for executing 600,000 instructions is
600,000 ÷ 30 million = 1 / 50 seconds = 20 milliseconds.
Therefore, the average processing time of a transaction is 20 + 60 = 80 milliseconds, which
means that d) is the correct answer.

Q2-42 a) State transition of a task

In control for multitasking, a task within a system is managed based on three states: the
“running state” is when the task in the system is being performed, the “ready state” is when the
task can be executed at any time, but its execution is awaited because the CPU is busy, and the
“waiting state” when I/O processing is being performed and its completion is awaited. A figure
that shows the transition of these three states is called a state transition diagram of tasks
(processes). When a task with a higher priority is set to the ready state, the task in the running
state is dispossessed of the right to use the CPU and is set to the ready state. Therefore, a) is the
correct answer. Such task switching (involving the appearance of a task with a high priority) is
called preemption.
b) A task that is generated is first set to the ready state.
c) When the processing of an I/O request ends, transition takes place from the waiting state to
the ready state.
d) When an I/O request is issued, transition takes place from the running state to the waiting
state in order to await the completion of the request.

42
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-43 c) Task execution pattern when the CPU idle time is zero

This question deals with the idle time when the CPU is not operating during the concurrent
execution of tasks. First of all, the CPU and I/O processing time of task A with a “high” priority is
expressed in the form of the following figure:

3 3 3 3 2
CPU and I/O of task A

CPU and I/O of a task

with a “low” priority (1) (2) (3) (4) (5)
3 3 or less 3 2 or less Any value

CPU operation I/O operation

Consideration should be given to filling the cells (1) through (5) so that the CPU idle time is
reduced to zero from the start of the execution of both tasks until the end of the execution of both
tasks. In order to reduce the CPU idle time to zero, the processing time for the CPU in (1) becomes
3, and the processing time for the CPU in (3) becomes 3. Since the idle time of I/O is not in
question, I/O time in (2) may be 3 or less and I/O time in (4) may be 2 or less. (5) is not related to
the idle time of the CPU, and can therefore be any value.
From the above conditions, task D described in c) that is executed in the sequence of
CPU(3) → I/O(3 or less) → CPU(3) → I/O(2 or less) → CPU(Any value)
is the correct answer.

43
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-44 c) Task scheduling algorithms

The scheduling algorithm by which processing is executed in order starting from the task (job)
with the shortest expected processing time is called the SJF (Shortest Job First) algorithm. This
algorithm is suitable for preferentially processing tasks that need to be handled immediately, such
as online real-time processing, however, when tasks with a short expected processing time arrive
in the CPU resource queue in quick succession, the tasks with a long expected processing time
wait indefinitely for the allocation of CPU resources, and might not be processed for a long time.
Therefore, c) is appropriate.
a) The scheduling algorithm where multiple CPU resource queues are created according to task
priority and CPU resources are allocated sequentially starting with the tasks with the highest
priority is called priority scheduling. The problem with this algorithm is that CPU resources
are not allocated to tasks with a lower priority very often. The process of gradually increasing
the priority according to the length of the waiting period so as to deal with this problem is
called aging.
b) The limited time period for which the CPU can be used is called a time quantum or time
slice. The scheduling algorithm by which the execution of a task is suspended once the
limited time period is reached and the task is appended to the end of the CPU resource queue
is called the round robin algorithm. The tasks in the CPU resource queue are processed in
sequence, and the allocation of CPU resources to the tasks is ensured.
d) This is an algorithm by which tasks are processed in order of arrival, and it is called FIFO
(First In First Out). Although the subsequent tasks are made to wait, once a certain amount of
time has elapsed, CPU resources are eventually allocated to the subsequent tasks as well.

44
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-45 c) Calculating the elapsed time of a job based on the scheduling algorithm

Since a time quantum is a significantly smaller value in comparison with the processing time of
a job, almost the same CPU time is allocated to jobs that are executed simultaneously. First, the
three jobs A through C are executed by allocating a time quantum with the round robin algorithm.
Therefore, approximately three times the amount of time taken to execute the jobs separately is
needed. As such, the time that elapses before the completion of job A (which has the shortest
processing time) is 5 (minutes) × 3 = 15 (minutes). After the completion of job A, job B and job C
are executed simultaneously. The time that elapses during this period is approximately twice the
time taken to execute the jobs separately. Therefore, the time taken until the completion of job B is
twice the “remaining time of job B” that is 5 (minutes) × 2 = 10 (minutes). Following this, job B
(that has the shortest processing time after job A) is completed, and finally, job C is executed by
itself.
In this question, the time that elapses before the completion of job B must be determined. This is
15 + 10 = 25 (minutes), and therefore, c) is the correct answer.

0 5 10 15 20 25 30 (minutes)

Job A Address

Job B

Job C

Job A Job B
Completed Completed

Q2-46 a) State in which processor utilization decreases

In a virtual memory system, if a program in the virtual memory is larger than the main memory
area that can be used in the system, or the number of programs running concurrently is large,
paging is performed very frequently. This is called thrashing as described in a) and causes the
efficiency of the entire system to decrease.
b) Fragmentation: This is a state in which the memory inside the storage device is fragmented and a
continuous area of the required size cannot be secured, and as a result, a series of memory areas
are arranged in a scattered manner. Frequent writing to the storage device and deletion of content
causes this state to progress, leading to a decline in system efficiency.

45
 Morning Exam Section 2 Computer System Answers and Explanations

c) Paging: In a virtual memory system, when no space is available in the real memory, the
programs in the real memory that are not necessary for the time being are pushed out to an
auxiliary storage device (page-out), and a program that might be necessary the next time is
imported into that area (page-in). These two operations are called paging. Pages that are
paged out are stored and managed at a location called a slot on the auxiliary storage device.
d) Bottleneck: It is a phenomenon in which the performance or capacity of an entire system is
severely limited by a resource with the highest amount of load, and which hinders
development in situations such as the development of a system.

Q2-47 a) Free space on the main memory in a variable partition system

The question states that “the modules are loaded from the beginning of a free space, and no
other conditions are considered.” Therefore, “Order of loading and unloading” can be considered
as follows:
(1) Load A → Load B → Load C (Loading the modules in order of A, B, and C from the beginning
of the main memory)
(2) Unload B → Load D (Loading D into the area from which B is unloaded as D can be loaded into
that area)
(3) Unload A → Load E (Loading E into the area from which A is unloaded as E can be loaded into
that area)

(1) (2) (3)

E
100 A 100 A 100

200 200 200

B D D
300 300 300

C C C
400 400 400

500 500 500

Free space

Loading from the beginning of the free space implies that if a module can be loaded, it is loaded
from the very beginning of the free space scattered throughout the main memory. If the area from
which a module has been unloaded is smaller than the size of the module to be loaded next, the
module cannot be loaded into that area. However, if the area is large, the module to be loaded next
can be loaded. Therefore, free space is available at three areas, and a) is the correct answer.

46
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-48 b) Effect of a virtual memory system on the main memory

A virtual memory system is a mechanism for providing an address space larger than the capacity
of the actual main memory by using an auxiliary storage device. Therefore, b) is the correct
answer. In a virtual memory system, a program is divided into several parts beforehand, and only
the parts necessary for processing are loaded to the main memory so as to enable the execution of
a program with a larger size than the actual memory capacity. The methods for the division of a
program include “segmentation” by which a program is divided into meaningful functional units,
and “paging” by which a program is divided simply into fixed units (such as 4 KB).
a) There is no effect on access speed to the main memory. To be exact, in some cases there is a
delay equal to the processing time for dynamic address translation. Furthermore, in the virtual
memory system, if the physical memory capacity is lower, the divided program is replaced
frequently, which increases the required processing time, thus resulting in a delay in the
effective access speed of the main memory.
c) Although in some cases the power consumption of the entire system increases by as much as
the amount of usage of the auxiliary storage device, there is no reduction in the power
consumption of the main memory.
d) Non-volatility is the ability to retain stored content even when the power is turned off.
Normally, auxiliary storage devices are non-volatile and the main memory is volatile, and
this holds true even for a virtual memory system.

Q2-49 d) Page replacement by the LRU algorithm

According to the LRU algorithm, the page that is not used for the longest time is paged out. In
this question, the state of page referencing and page replacement is arranged in a table in view of
the fact that “no page exists in the main memory in the initial state”. As a result, it is clear that
when page 6 is accessed at the end, it is stored at the position where page 5 was stored. Therefore,
d) is the correct answer.

1 2 3 4 5 6 7 8 9 10
1 1 1 1 1 5 5 5 5 (5) 6
2 2 2 2 2 2 2 2 2 2
3 3 3 3 3 1 1 1 1
4 4 4 4 4 3 3 3

Vertical axis: Page frame Horizontal axis: Access frequency

47
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-50 c) Explanation of the spooling function

The spooling (spool) function refers to the I/O process of data using a spool area. The spool area
is the storage area for an auxiliary storage device that plays the role of a buffer for releasing the
CPU from the I/O operation of a low-speed device. For example, the data to be output to a printer
is first stored on a high-speed hard disk device (spool area), and then output to the target printer. In
this way, the program can be ended when the data is written to the spool area, and another process
can be started. By quickly separating the program from a low-speed I/O device such as a printer,
the number of programs that can be executed concurrently can be increased and the low-speed I/O
device and CPU can be used effectively, which improves the throughput of the entire system.
Therefore, c) is the correct answer.
a) This is a description concerning scheduling of tasks.
b) This is a description concerning interrupt.
d) This is a description concerning buffering that is used when a disk is accessed. A disk cache
also achieves a comparable function.

Q2-51 c) Calculating the number of a page fault

According to the conditions in the question, a single access to the main memory is performed in
200 nanoseconds, and the overhead for a single page fault is 100 milliseconds. Furthermore, a
page fault occurs once every 500,000 accesses to the main memory. Based on this, if the average
time required to access the main memory 500,000 times is calculated by including the overhead
for the one page fault that is expected to occur during this time, it is calculated as the sum total of
−1
the time required for 500,000 accesses and the overhead due to the page fault, which is 2 × 10
(seconds), as shown in the figure below:

Time required for accessing the main memory 500,000 times Page fault
4 −9 −1 −1
50×10 × 200 × 10 = 10 (seconds) 100 milliseconds = 10 (seconds)

The numeric value to be obtained is the maximum number of page faults per second, but since a
page fault occurs every 500,000 accesses to main memory, the time period necessary for 500,000
accesses (including the page fault), which was calculated earlier, is used as a unit to determine the
maximum frequency of occurrence in one second. Thus, the maximum number of the page faults
−1
per second becomes 1 ÷ (2 × 10 ) = 5 (times/second), which means that c) is the correct answer.

48
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-52 c) Printing process

By using an auxiliary storage device as buffer storage in order to bridge the difference in the
processing speed of a processor and peripheral devices, storage of the input from the peripheral
devices or the output to the peripheral devices in an area inside the auxiliary storage device is
called spool control. Furthermore, the file used to store information about input and output in order
to perform this spool control is called a spool file.
In the question, point (3) under “Conditions” states that a spool file for printing with a size of
400 MB is secured for the execution of a single job. Furthermore, points (4) and (5) under
“Conditions” state that after the execution of a job, the contents of the spool file are processed by
the printing function of the OS, and that once printing is completed, the spool file is deleted. Here,
the deletion time is ignored. That is, a spool file of 400 MB is required from the time of execution
of the job until printing of the print data (the output from the job) with the printing function of the
OS is completed.
Here, as stated in point (2) of “Conditions”, the execution time of one job is 20 minutes.
Furthermore, as stated in point (6) of “Conditions”, the printing time is 10 minutes for every 100
MB (= 10 MB/minute), and since the size of the spool file is 400 MB, the printing time is
calculated with the following expression:
400 (MB) ÷ 10 (MB / minute) = 40 (minutes)
According to point (1) of “Conditions”, a job is executed continuously four times with a
multiplicity of 1, and according to point (6) of “Conditions” one printer is available. Finally,
according to point (7) of “Conditions”], the jobs can be executed and printed in parallel, and these
processes do not affect each other. Therefore, if the execution and printing process of the jobs is
compiled in chronological order, it appears as shown below.

20 minutes 40 minutes

Job (1) Job (1)

Execution Printing
400M Waiting
minutes
for 40 minutes
20 minutes printing
Job (2) minutes Job (2)
Execution Printing
400M
minutes Waiting for
20 minutes printing 40 minutes
Job (3) minutes Job (3)
Execution Printing
400M
minutes
Waiting for
20 minutes printing minutes 40 minutes
Job (4) Job (4)
Execution Printing
400M
minutes

0 20 40 60 80 100 120 140 160 180 [minutes]

49
 Morning Exam Section 2 Computer System Answers and Explanations

By 60 minutes from the start of execution of the first job, three spool files of 400 MB each are
necessary. However, since the first spool file is deleted after 60 minutes, 400 megabytes need not
be secured afresh for the fourth spool file. Thus, three spool files of 400 MB each are required
between 40 minutes and 100 minutes after the start of the first job and their capacity becomes
1,200 bytes. Therefore, c) is the correct answer.

Q2-53 b) Explanation of API in an OS

API (Application Program Interface) is a mechanism for using various functions of the OS from
application software, and is a collective term for interfaces, such as the functions, commands, and
utilities provided for easy development of a program. Therefore, the description b) is appropriate.
a) This is an explanation of a mechanism that is used to operate the hardware directly from an
application without using APIs (such as Windows DirectX), in order to improve the speed of
image rendering.
c) The mechanism of communication between multiple applications across a network includes a
socket, which is a TCP/IP-based communication interface between processes.
d) This is an explanation concerning the integration of GUIs (Graphical User Interfaces), which
includes Motif and OPEN LOOK of UNIX.

Q2-54 b) Explanation of middleware

Middleware refers to software that functions in an intermediate position between the operating
system and application software and provides generalized versions of functions that are commonly
used by the application software. It also performs the role of effectively linking the application
software and the operating system. Therefore, b) is the appropriate explanation.
a) This is an explanation of the operating system. It manages various resources such as the
hardware and data, and ensures their effective use. Besides control programs in a narrow
definition of an OS, language processors and service programs are also included.
c) This is an explanation of an ERP (Enterprise Resource Planning) package. An ERP package
is a business package that integrates the mission-critical business systems of a company. It
can support wide-spread activities from the procurement of raw material to production, sales,
and personnel affairs.
d) This is an explanation of groupware. Groupware is software that supports a mechanism for
collaborative work by a group. It has functions for sharing document information, providing
communication tools (such as e-mail and electronic bulletin board), and managing workflow.
Q2-56d)

50
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-55 d) Moving the current directory

A file system that has a hierarchical structure is composed of the root directory itself and files
and directories belonging to the root. Files and directories can be specified with an absolute path,
which is the path from the root, or by a relative path, which is the path from the directory being
handled currently (current directory). As mentioned in the question, when the current directory is
moved, the absolute path of the last current directory is to be determined.
• \A\B
This is an absolute path. A is the directory immediately below the root, and then B is under
that, so the current directory is (1).
• ..
This means the directory that is one level above. Therefore, the current directory is (2).
• ..\B
Two periods means one level above (2), which is the root. B is under the root. Therefore, the
current directory is (3).
• .\A
The current directory is B, which is (3), and the directory A under (3) is the next current
directory. Therefore, the last current directory is A, which is (4).
The absolute path of A corresponding to (4) is to be determined. The absolute path of A begins
with the root (\) followed by B which is immediately below (\B), and then reaches A that is below
B. Therefore, the absolute path is \B\A as described in d).

Absolute path \A ... a) Root(\)

(3)
A (2) B
Absolute path
\A\B\A ... c)
Absolute path \B\A ... d)
(1) (4)
A B First A Last B
Absolute path
\A\A ... b)
A A B

51
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-56 d) Explanation of archive

In English, archive means a record, storage, or old document. In computer terminology, archive
means to compile multiple files into a single file, and store this on a storage device, as described in
d). In many cases, in order to effectively use the storage media, or improve the efficiency of
backup and distribution, the data is compressed. A file compiled as an archive is sometimes called
a stack.
a) This is an explanation of “data save”.
b) This is an explanation of file mirroring.
c) This is an explanation of the update log of the database.

Q2-57 c) Explanation concerning a language processor

Language processor is a collective term for software that performs processing concerning
programs, such as translating the source program coded in a programming language and creating
an executable program. After reading the description of each of the options, the description of a
generator provided in c) is appropriate.
a) An assembler is software for creating a machine language program by translating the source
program written in a programming language (assembler) with a one-to-one correspondence to
machine language. The software that converts a source program into a source program for
another processor is called a converter.
b) An interpreter is software that “executes a program written in a high-level language while
interpreting it on a per line basis” as described in d). The description “A microprogram that
decodes and executes the program intended for another computer” indicates an emulator.
Furthermore, a microprogram is a program installed in a CPU that processes complex
machine language instructions like software.
c) A generator is software that creates a program as appropriate for the purpose of processing by
specifying the necessary conditions as parameters, and is the correct description. Among well
known generators, RPG (Report Program Generator) is popular. In the recent years, products
that generate HTML from created GUI images and those that generate a source program from
a structured chart have appeared, and these can also be classified as generators.
d) A translator is software that converts a source program written in a high-level language into a
source program in another high-level language.

52
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-58 c) Explanation of optimization in a compiler

Optimization of a compiler implies changing a program so as to reduce the execution time and
size of the object program without changing the functions of the program. Therefore, c) is
appropriate. For example, deletion of redundant instructions, and reorganization of loops and
register assignment are performed. Optimization is mostly performed after semantic analysis
within the flow of a series of compilation processes.
(Lexical analysis → Syntax analysis → Semantic analysis
→ Code optimization → Code generation)
a) This is a description concerning an interpreter. An interpreter executes a program while
translating it, and therefore, it does not generate an object code.
b) This is a description concerning a cross compiler.
d) This is a description concerning a function called snapshot dump that is used for debugging.
If the debug mode is set as an option for compilation, most compilers usually generate such
object code.

Q2-59 a) Functions of a programming tool

An inspector is a dynamic analysis tool that displays the content of the data during the execution
of a program for the purpose of debugging. Therefore, a) is correct.
b) A simulator is a tool for simulation and includes terminal simulators that simulate the
operation of a specific terminal, and traffic simulators that generate a specific communication
pattern. A tool that displays the execution path within a program or between programs is
called a tracer.
c) As described earlier, a tracer is a dynamic analysis tool. In order to simplify the search for
functional explanations and data definitions in each program, repositories are used.
Repositories are databases for managing the results all together including information
described above.
d) A browser is Web client software for displaying Web pages. A tool that edits the source code
of a program based on functions, such as character insertion, deletion, and substitution, is
called an editor.

Q2-60 a) Flow from interpretation of a program until its execution

(1) A compiler interprets the source program written in the language of that compiler.
(2) When the compiler interprets the source program, it becomes “A: object program” in machine
language or an intermediate language.
(3) A linker (linkage editor or linkage editing program) performs linkage editing of “B: library”
module from several object programs and program libraries, and creates a “C: load module”
(executable program). The address at this stage could also be a relative address (depending on
the type of computer) so that it can be allocated at any location in the main memory.

53
 Morning Exam Section 2 Computer System Answers and Explanations

(4) A loader loads a load module into main memory while converting it to a real address and
executes it. In current OSs, the OS itself has a function for loading programs (load modules)
to main memory, so an individual program such as a loader is not needed.
Therefore, a) is appropriate.

Q2-61 d) Open source software defined by OSI

OSI (Open Source Initiative) is a non-profit organization that promotes open source. Open
source is a software license that enables free browsing of the source code, as well as change and
redistribution of the content while protecting the copyright of the software. Software based on this
license is called open source software. According to the definition of open source software by OSI,
10 criteria must be fulfilled as conditions for distributing a program. Based on this, when the
modified software is redistributed, the source code must be released to the public, however, for
private or in-house use and modification, the source code does not need to be released to the
public. Therefore, d) is the correct answer.
a) The use of a program cannot be restricted to only a specific field.
b) When the open source software is modified and redistributed, its distribution under the same
license as the original software must be allowed. However, this does not imply that the
modified open source software must be distributed under the same license.
c) The software developer cannot restrict the sale or free distribution of the open source
software by a third party, and cannot ask for royalties or other remuneration concerning sales.

Q2-62 d) Description concerning the flash memory

A flash memory is an EEPROM (Electrically Erasable Programmable ROM) that can be erased
all at once, and is used as the storage media in digital cameras and IC cards. In terms of
semiconductor memory, a rewritable ROM (Read Only Memory) is called PROM, and among
PROM, a memory that can electrically delete and write content in each block is called a flash
memory (flash EEPROM), which has a property known as non-volatility in that its content cannot
be deleted even when power supply is turned off. Therefore, d) is appropriate.
a) This is a description concerning SRAM (Static RAM). A refresh operation is not necessary
and data can be read fast, but the cost is higher than that of DRAM.
b) This is a description concerning EPROM (Erasable Programmable ROM). It is a type of
PROM whose content can be erased by ultraviolet radiation. It is mostly used for storing
programs.
c) This is a description concerning DRAM (Dynamic RAM). It is a semiconductor memory
with a simple structure and a high degree of integration, which requires data to be refreshed
so that the recorded content is not deleted.

54
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-63 b) Logic circuit from which the same output is obtained

The given logic circuit is X = ( A • B ) + B , which appears as below when depicted with a Venn
diagram:

A B A B A B
＋

A• B B ( A • B) + B

This can also be considered as the logical sum of “A” and “not B”.
Next, the logic circuits of the options are evaluated as below, and the same output results are
obtained from the circuit shown in b).
a) X = A• B
b) X = A+ B
c) X = A + ( A • B) = A
d) X = A • B = A + B = A + B (The form is changed using the De Morgan’s Law)
For reference, the double negation of De Morgan’s Laws can be used to prove this as follows:

X = A • B + B = A • B + B = ( A • B) • B

= ( A + B) • B = A • B + B • B = A • B = A + B

Q2-64 a) Combination of gate devices in a half adder

As this is a half adder, the gate devices can be obtained simply by adding the binary numbers.
When x + y is expressed as a binary number,

0 0 1 1 ← x
+0 +1 +0 +1 ← y
00 01 01 10 ← cz (carry bit and first digit of the sum)

z is calculated with the exclusive logical sum because it becomes 1 when either input is 1.
c is calculated with the logical product because it only becomes 1 in the case of 1 + 1.
Therefore, the combination of a) is appropriate.

55
 Morning Exam Section 2 Computer System Answers and Explanations

Q2-65 c) SoC

SoC (System on a Chip) is a system LSI in which several functional circuits are incorporated
into a single semiconductor chip. Therefore, c) is appropriate. Although the development cost is
higher than the SiP of d), it is suitable for compact products that are in mass production.
a) This is a description concerning an electronic circuit board such as the motherboard of a PC.
b) This is a description concerning the chipset of a PC. For example, the Intel 855 chipset by
Intel Corporation is a chipset for a notebook PC.
d) This is a description concerning SiP (System in a Package) in which multiple chips are stored
in a single package. System LSI is broadly divided into SoC and SiP. For reference, system
LSI is a general-purpose LSI, and custom LSI indicates an LSI that contains the specific
functions desired by the user. Custom LSI is divided into a special-purpose ASIC
(Application Specific Integrated Circuit) and a programmable FPGA (Field Programmable
Gate Array).

56
 Morning Exam Section 3 Technology Elements Answers and Explanations

Section 3 Technology Elements

Q3-1 d) Accessibility

Accessibility is a keyword that refers to the ease of using information systems, such as software,
information services, and Web sites, on a user interface. Therefore, d) is the correct answer. The
color combination of a screen, design including font size, provision of audio and speech functions,
and the attachment of text descriptions of images and audio so that everybody including elderly
people and people with disabilities can also adequately access the information systems is also “the
degree of achieving universal design in information systems.” In addition, not only items
standardized with barrier free design and functions, but it also includes equipment and software
that allows independent setting and adjustment of these functions according to the physical
characteristics of the user.
a) This refers to the basic resident register network system. By connecting the administrative
agencies and centers all over the country with local public agencies (prefectural and city
governments, municipalities), this system makes it possible to share and use information
about specific users to whom a resident register code is assigned (recorded in the basic
resident register).
b) This refers to interoperability.
c) This refers to traceability.

Q3-2 c) User interface

A user interface that allows frequently-used operations to be performed efficiently is a shortcut

key. Frequently-used operations are assigned to keys beforehand, and instead of entering a
command or operating a GUI, an operation is performed by simply pressing the corresponding
key. One of the main examples is, pressing [Ctrl] and [C] simultaneously performs the copy
operation. Therefore, c) is the correct answer.
a) Undo function (returning to the original) is a user interface for returning to the state prior to
making a change.
b) Online help is a user interface used when the user is not familiar with the operation method.
d) Progress bar is a GUI indicating the progress of a process.

57
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-3 b) Input method for GUI screen

Input methods for GUI screen include text box, pop-up menu, list box, radio button, checkbox
and the like.
A text box is a method of directly entering characters. It includes an input mode where the
maximum number of characters allowed in the relevant fields is set and no restrictions are placed
on the input characters, and an input mode where characters allowed in the relevant field are
decided, and the entering of any other characters results in an error message and user is asked to
re-enter the characters.
A pop-up menu is a menu displayed at any position on the screen, and a user selects his or her
intended item from the menu.
In a list box, options are displayed inside a box. An items list is displayed vertically and the user
searches for and selects his or her intended item by using a scroll bar on the right side of the list
box. The selected item is highlighted by inverting the brightness and colors of the display, and
once the input is completed, the highlighted item is entered.
Radio buttons are used when the user is required to select only one item from multiple items
displayed beforehand.
A checkbox is used when the user selects as many items as needed from multiple items
displayed beforehand.
For the “method of choosing from the list of options” in this question, the possible choices
could be pop-up menu, list box, radio button, and checkbox. When there are only a few types of
input data, and the content of data is fixed, these methods are effective. Therefore, b) is the correct
answer.
a) As the method in question here is for selecting items, basically the items cannot be modified.
When it is necessary to modify each item, a text box with default setting (initial value) is
appropriate.
c) When the input data can take many values, it is cumbersome to search for the value to be
selected. Therefore, in most of such cases, a text box is used where the value can be directly
entered.
d) In order to enter a large amount of data such as sentences, a text box is the appropriate
method.

58
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-4 b) Methods for checking data

In balance check, individual tabulation is done with respect to a pair of data items where the
final sum total should be balanced, and then it is checked that the difference between total values
is zero. Therefore, b) is the appropriate answer.
a) In the method of using check digits, one digit is appended for the detection of input errors in
code. The value of this one digit to be appended is obtained by performing certain calculations
on the original code. After that, it is continuously checked whether this relation still holds, and
in that way, errors are detected.
c) A format check is for checking the format of an input field, and not an input item. An input
format (categories such as alphabet letters, numeric value, and double-byte characters) is
defined for each item, and it is checked whether the actually entered value matches the
predefined format. For example, entering a numeric value in the field where only alphabetical
characters are allowed results in an error.
d) A limit check is for checking whether a value entered in a field is within the permissible limit
or not. For example, entering $200 in the field where the value must not exceed $100 results
in an error.

Q3-5 b) Design principles of input screen

As input screens are screens for entering information that is important for the information
system, it is desired to have design principles that help in designing screen layout that is easy to
understand and easy to use for users, and that also prevents input errors. Therefore, b) is the
correct answer.
a) In the case of a wrong input, if a program terminates without being able to perform
subsequent processes, users cannot use it comfortably.
c) As users enter data by looking at the input form, layout designed according to the input form
is easier to use.
d) Even if users are proficient with the operation because of frequent use, an error may occur
accidentally as long as it is a manual input, and therefore input check cannot be omitted.
Moreover, it should be kept in mind that input errors occur until the user becomes proficient
with the operation.

59
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-6 c) Characteritics of various codes

Block code, group classification code, and sequence code that appear in the answer choices have
the characteristics listed below respectively.
· Block code: A range of code values is decided for each group, and code values are assigned
within this range.
· Group classification code: In this code, each digit has a meaning of code, and in most of the
cases, starting in sequence from the top, digits have the meaning of major category,
medium category, minor category, and serial number. The typical group classification
code is student numbers that contain year of enrollment, department, academic discipline,
and roll number.
· Sequence code: As indicated by its name, code names are assigned in the sequence of
occurrence.
In addition to these three codes, like product code of electrical appliances, mnemonic code is
also a major code that assigns values so that a product can be associated with the code itself.
a) As one of the advantages mentioned here is “codes in the order of occurrence are assigned,” it
indicates sequence code. As this code assigns code values without leaving any interval in
between, the number of digits is fewer. However, this code does not consider categories, and
therefore it has the characteristics described in disadvantages and application area.
c) As one of the advantages mentioned here is “each digit has a specific meaning of
classifications,” it indicates group classification code. Other characteristics of this code are,
the number of digits can easily become large as mentioned under disadvantages, and it can be
used when the classification criteria is clear.
As the meaning of the aforementioned two choices is known, it is clear that the correct answer is
c).
b) Although the number of digits is fewer compared with the group classification code, block
code suffers from the disadvantage that codes cannot be assigned when digits become
inadequate in the range allocated to the group. Moreover, in the group classification code, the
number of digits is decided from the standpoint of categories. However, in the block code, the
range of codes is partitioned by each group, and therefore even if there are restrictions on the
number of digits, grouping is possible.

Q3-7 d) Explanation of usability of Web contents

Usability refers to “ease of use.” The concept of usability of is not only for people with
disabilities or elderly people, but it is a term that expresses general ease of use in a broader sense.
Therefore, d) is appropriate for the usability of Web contents. In addition, in the international
standard ISO 9241-11, “Guidance on usability” is provided as a principle for ergonomically
measuring the ease of use of visual display terminals in offices. In this guidance, usability is
defined as “the extent to which a product can be used by specified users to achieve specified goals
with effectiveness, efficiency and satisfaction in a specified context of use.”
a) This is the concept of universal design.
b), c) Both are the concepts of barrier free design.

60
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-8 c) Capacity of audio sampling

Sampling can be considered as “picking up samples.” Audio sampling is a technique of

encoding analog data such as audio (that is, converting it into digital format), and fetching the size
(amplitude) of a wave at a certain interval from the continuous analog waveform.
Data is sampled 11,000 times per second, and each value is recorded in 8-bits. Based on the fact
of “8 bits = 1 byte”, the amount of data sampled in a second is calculated as follows:
8 bits/time × 11,000 times/second = 88,000 (bits/second) = 11,000 (bytes/second)
= 11×103 (bytes/second)
Therefore, in a flash memory with a capacity of 512×106 bytes, audio data of
(512 × 106 (bytes)) ÷ (11 × 103 (bytes/second)) = 512 × 103 ÷ 11 (seconds)
= 46545.45 (seconds),
46545 ÷ 60 =775.75 (minutes) can be recorded.
Therefore, c) 775 (minutes) is the correct answer.

Q3-9 c) International standard of compression encoding of still image data

When image or audio data is handled in a multiple processing system, the data volume is very
large. Therefore, data compression techniques are required. JPEG (Joint Photographic Experts
Group) and GIF (Graphics Interchange Format) are the main compression methods for still image
information. JPEG is an international standard defined by ISO/IEC. Therefore, c) is the correct
answer.
There are two types of JPEG compression methods: lossless compression and lossy
compression. In the “lossless” method, when the compressed image data is restored to its original
form, it can be fully restored, while in the “lossy” method, the compressed image data cannot be
fully restored. Although the compressed image data cannot be fully restored, a human eye hardly
feels the difference.
a) BMP (Bit MaP) is used as a standard graphic format used by Windows. Color and shape data
is managed in units of bits. Screen is clear and sharp as data is not compressed, but file size
becomes large.
b) GIF (Graphics Interchange Format) is used as a compressed image format of the lossless
compression method. LZW method is used for compression, and the more repeated parts of
the same color or pattern, the higher the compression ratio. Therefore, it is more suitable for
compressing illustrations. However, it is not yet standardized by international organizations
such as ISO.
d) MPEG (Moving Picture Experts Group) is used as an international standard of compression
technology for moving images.

61
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-10 b) Features of compression methods of still image data

There are two data compression methods: lossless encoding (lossless compression) and lossy encoding
(lossy compression). In the lossless encoding method, data before compression and data after
compression and decompression process completely matches. In this method, data after decompression
completely returns to its original form. Therefore, image quality is constant irrespective of the
compression ratio. Therefore, description of b) is appropriate.
On the other hand, in the lossy encoding method, data before compression and data after compression
and decompression process does not match completely. Although the quality of data after decompression
is reduced, a high compression ratio can be easily achieved. Deterioration in quality is made less
noticeable by using audio-visual characteristics, and accordingly it is used for compressing general image
and audio data. In JPEG, which is one of the typical still image compression methods, although the lossy
encoding method is mainly used, the lossless encoding method is also included as a standard.
a) The compression ratio is lower in the lossless encoding method in comparison with the lossy
encoding method, and the file size becomes larger.
c) In the lossy encoding method as well, the image size does not become smaller. In most cases,
the size is kept the same while the resolution is reduced.
d) In the lossy encoding method, it is usual that the compression ratio can be changed. However,
there is a trade-off between quality and compression ratio.

Q3-11 b) Explanation of virtual reality

Virtual Reality (VR) is also known as artificial reality. It refers to “information mediated by
computers is used for having people experience and feel as though it is the real world.” In virtual
reality, objects and space simulated with a computer are displayed by using computer graphics
such that they create a perception of the real world.
As described in b), a world created in computers by using technologies such as CG and sensor is
presented as the real world. I/O devices, such as an HMD (Head Mounted Display) that provides a
stereoscopic view so that objects appear stereoscopic and a data glove that conveys the sensation
of touching the objects, are used. It is used in education, training, medical treatment, and design.
a) This describes “progressive JPEG” (Joint Photographic Experts Group) and “interlace GIF”
(Graphics Interchange Format), which is one of the main image file format used on the
Internet. When common GIF format image data is displayed, it gradually displays from the
top to the bottom of the image while it is downloaded. However, in the interlace GIF, the
entire screen first appears in mosaic form, and it gradually becomes clear as the download
progresses. Interlace means to weave, and it originates from alternatively moving the scanning
lines of display. Moreover, progressive JPEG is also a similar method of display, where even
if the download is not complete, the view can get glimpses of the entire image.
c) This describes simulation, which is one of the applications of virtual reality, but it is not
appropriate as an explanation of virtual reality itself.

62
 Morning Exam Section 3 Technology Elements Answers and Explanations

d) This refers to VFX (Visual Effects), and it indicates the technique of synthesizing a real
recorded subject matter by using computers. SFX (Special Effects) is similar or synonymous
to this, but it refers to synthesis technique that does not use computers, and in other words, it
is “special photographing.” In some cases, no distinction is made between VFX and SFX.

Q3-12 a) Computer graphics

When a shape in computer graphics is drawn, technology used for making the step-like jagged
appearance near edges less noticeable is anti-aliasing. An image is displayed by changing the color
of pixels constituting the image. However, pixels can only display the color corresponding to the
assigned numeric value, and therefore the number of colors that can be simultaneously displayed is
only one at maximum. Therefore, a boundary of two colors has a step-like jagged appearance, and
it is necessary to assign intermediate color to make is less noticeable. This process is called
anti-aliasing. Therefore, a) is the correct answer.
b) Clipping refers to specifying a portion of an image for restricting the processing range.
c) Shading refers to assigning shade for giving a three-dimensional appearance.
d) Morphing is creating an intermediate image from images before and after the change so that
the image changes smoothly.

Q3-13 d) Explanation of relational database

Database models are broadly divided into a relational model and structured model, and the
structured model is further classified into a hierarchical model and network model. While the
parent-child relationship in a hierarchical model is “1 : n”, the parent-child relationship in a
network model is “n : m” and it is also called the CODASYL model. Relational model represents
the collection of data in several two-dimensional tables, and relation is represented by associating
key items between these tables. Therefore, d) is the appropriate explanation.
a) A hierarchical model database shows relation between data and the hierarchical structure.
b) A list links the related data by using pointers, and rather than database, it is one of the
methods of showing linked structures of data.
c) An object oriented database (OODB) is formed by combining data and its operation procedure
in the form of a database.

63
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-14 b) Three-schema architecture of database

The three-layer schema architecture is composed of an external schema (a group of data items
describing the required logical structure as seen from the viewpoint of an individual user), a
conceptual schema (a group of data items describing the logical relation to be originally contained
in the data), and an internal schema (a group of data items written in view of hardware,
performance, recovery, and security). The relation between these three is shown in the figure
below.

Internal schema

Conceptual schema

External schema External schema External schema

... ...

Logical data independence is achieved by separating the conceptual schema (which shows the
logical relation of data) from the external schema (which shows the view desired by users). In
specific terms, even when there is a change in the logical relation of data (addition or deletion of
tables or data items), if this change is not related to the view of data desired by users, the schema is
not affected by this change. Therefore, b) is the appropriate description.
a) As mentioned above, the three-layer schema architecture refers to external schema,
conceptual schema, and internal schema.
c) (Incorrect) Internal schema → (Correct) Conceptual schema
d) (Incorrect) External schema → (Correct) Internal schema

64
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-15 b) Sequence of data entry in table

This question is about referential constraints in a database. Referential constraint refers to the
constraint that “when a value in a column of some table refers to another table, a column in the
referenced table must also have that value.” In a relational database, the foreign key (an item that
is used for association with another table, and that is the primary key of the referenced table) must
maintain referential constraint with the primary key of the referenced table.
Here, first it is necessary to understand from the data items of a table how referential constraints
are set. In the “Asset” table, PC_number, Building_number, and Room_number are the primary
keys. It is required to check whether or not the combination of Building_number and
Room_number actually exists in the “Room” table. Moreover, it is also required to check whether
or not Building_number in the “Room” table exists in the “Building” table. Summarizing this
gives the result shown below.

Asset PC_number Building_number Room_number Model_name

Room Building_number Room_number Room_name

Building Building_number Building_name

As the referenced data must be entered first, sequence of data entry in the tables is Building →
Room → Assets. Therefore, b) is the appropriate answer.

Q3-16 c) Primary key of a relational database

The field required for identifying each row is called the primary key, and in the same table no
two rows have the same primary key. Therefore, c) is the appropriate answer. In other words,
columns that have rows with the same primary value cannot be a primary key.
a) Rows can be searched even if search conditions are specified for items other than the column
identified as a primary key.
b) Arithmetic operations can be performed on any column irrespective of whether it is a primary
key or not.
d) Multiple columns can be specified as a primary key. In this case, a primary key is also called a
composite key.

65
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-17 c) Third normal form

The outline of procedures until achieving the third normal form is given below.
- First normal form: Separate the repeated parts.
- Second normal form: Separate the attributes that are functionally dependent on the entire
composite key (full functional dependency), and that are functionally dependent on only a part
of the composite key.
- Third normal form: Other than key data, search the data that can become a key, and separate it.
Applying the tasks involved in this procedure to the answer choices, it is clear that the table in
c) is in the third normal form.
a) Employee_name is functionally dependent on Employee_number only, while Skill_name is
functionally dependent on Skill_code only. Therefore, there is no relation between employee
and skill.
b) Employee_name is functionally dependent on Employee_number only, and just like a),
conditions of the second normal form are not satisfied. In this situation, when one employee
has multiple skills, the same combination of Employee_number and Employee_name is
shown redundantly.
d) There is only one item of Skill_years_of_experience for each employee, and years of
experience for each skill cannot be recorded.

Q3-18 d) Identifying the primary key from functional dependency

A primary key of a relation is a minimum set of attributes with which a tuple of a relation
(called a row in SQL) can be uniquely identified. Minimum means that even if one attribute is
removed from that set of attributes, it doesn’t fulfill the role of a primary key. In other words, it
indicates the minimal requirement.
a) From the details of functional dependency of (1) through (7), it is clear that product number,
product name, or quantity cannot be decided from order number only. Therefore,
(Order_number) is not the primary key.
b) Just like a), product number, product name, or quantity cannot be decided. Moreover, in the
functional attribute (2), there is Order_number→Customer_number. Therefore,
Customer_number can be decided by Order_number, so it does not make sense to add
Customer_number (excess or redundant). Although (Order_number, Customer_number) can
uniquely identify a tuple, this combination is not minimum. It is clear that it is not the primary
key.
c) With these three attributes, all other attributes can be decided. However, just like b), as it
includes Customer_number, it is not minimum, and it cannot be the primary key.
d) According to the functional dependencies of (1) through (7), all attributes can be decided with
(Order_number, Product_number) (identification and equivalent of tuple), and it is minimum.
Therefore, it matches the definition of primary key, and d) is the correct answer. Moreover, by
comparing with c), it should be kept in mind that the required quality of a primary key is
“minimum.”

66
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-19 a) Operation of a relational database

In the operation of a relational database, retrieving the respective rows that satisfy the specified
conditions from multiples tables, and then creating a new table (derived table) is called “Join”.
Therefore, a) is appropriate.
b) Project – Retrieving columns (not rows) that meet certain conditions from a table.
c) Select – Retrieving rows (not columns) that meet certain conditions from a table.
d) Insert – Inserting specific rows (not columns) in a table.

Q3-20 a) Data manipulation not subject to referential constraints

Referential constraints are constraints concerning the integrity of reference relationship between
tables in a relational database. In a relational database, for referring to an item in another table by
using a foreign key, the referring item must exist in the referenced table.
In the three tables provided in the question, items referring to other tables as a foreign key
(items indicated with a dotted underline) are Product_code and Customer_code in the Order table,
and they refer to the Product table and the Customer table respectively.
a) The “Customer” table is the referenced table, and new records can be added to this table
without any constraints.
b) The “Product” table is the referenced table, and when the records whose product codes exist
in the “Order” table are deleted, constraints are applied.
c) The “Product” table is the referenced table, and when product codes to be changed exist in the
“Order” table, constraints are applied.
d) As Product_code and Customer_code of the record to be newly added to the “Order” table
must already be registered in the “Product” table and the “Customer” table respectively,
constraints are applied.
Therefore, a) is the correct answer.

Q3-21 c) Information that cannot be obtained from data

As one buyer may purchase multiple products, from “Daily sales data of sales agents,” “Daily
variation in the number of buyers for each sales agent” cannot be obtained. Moreover, as the Date
field is not present in “Product data bought by customers,” “Daily variation in the number of
buyers for each sales agent” cannot be obtained. As both “Daily sales data of sales agents” and
“Product data bought by customers” have Sales_agenct as a common field, it is possible to join the
records in these tables where the Sales_agenct field has the same value. However, even after the
daily sales data and product data bought by customers are joined, it is not possible to determine the
daily number of buyers. Therefore, c) is the correct answer.
a) “Daily variation in sales quantity for each product” can be obtained from Date, Product, and
Sales_quantity in “Daily sales data of sales agents.”

67
 Morning Exam Section 3 Technology Elements Answers and Explanations

b) A single record can be created by joining the records in “Customer data” and “Product data
bought by customers” that have the same value in the Customer field. Therefore, “Hot-selling
products by gender” can be obtained from Gender, Product, and Sales_quantity.
d) “Age distribution of the buyers for each sales agent” can be obtained from Customer,
Date_of_birth, and Sales_agent.

Q3-22 b) Description concerning SQL statements

Verifying the results of SQL statements a) through d) including set functions provided in the
questions gives the following outcome.
a) AVG is a function for calculating a mean value. Data in the first and the third record is
applicable for Product_number NP200 specified in the selection condition, and calculating the
mean value for this quantity gives (3+1)÷2=2.
b) COUNT is a function for calculating the number of applicable data records. As no specific
selection condition is specified, all records are counted. Therefore, the outcome is 4.
c) MAX is a function for calculating the maximum value. Calculating the maximum value of
quantity gives 3.
d) SUM is a function for calculating the total value. Data in the third and the fourth record is
applicable for date 2006-10-11 specified in the selection condition. Calculating the total
quantity for these records gives 1+2=3.
Based on the above results, the maximum value is obtained in b).

Q3-23 b) Correct SQL statement

When there is a GROUP BY clause in the inquiry, the selection list of the SELECT clause must
be a column (arithmetic expression including that column) specified in the GROUP BY clause, or it
must be a constant number or a set function. In other words, if there is a GROUP BY clause, only
values representing a group can be specified in the SELECT clause. b) satisfies this condition, and
it is the correct answer.
a) Although set function is specified in the selection list of the SELECT clause, GROUP BY
clause is not included in this inquiry. Therefore, this can be considered as one group. Set
function of a value representing one (1) group is allowed, however, Order_date is not a value
(one in this case) representing a group, and hence it cannot be allowed.
c) There is an error here as the set function is nested.
d) Since a set function is included in comparison predicate of the WHERE clause, there is an error.
For specifying a set function, use the HAVING clause.

68
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-24 a) Selection of SQL statement

As this “SQL statement is used to search for Department_code of departments that have less
than five employees with the job duty of Programmer,” the retrieval conditions are as follows:
(1) Employees with the job duty of programmer
(2) Departments that have less than five employees falling under (1)
Normally, an SQL statement that joins the “Department” table and the “Employee” table by
using Department_code can be considered, and then employees with a Job_duty of Programmer
can be retrieved. However, only the “Employee” table is provided in the question statement.
Therefore, as given in answer choices, let the “Employee” tables be S1 and S2, and S2 be the
correlated subquery to be referred for one row of S1.
Here, “departments that have less than five employees with the job duty of Programmer” means
that not only those departments that have one to four employees as programmer, departments with
no employees as a programmer should also be considered.
a) S2 is referred to for one row of S1, and the number of rows of Employee_number that have
the same Department_code and Job_duty of Programmer are tabulated with the COUNT
function. As a result, the Department_code of rows that have less than five (including zero
(0)) as an outcome is retrieved from S1. However, as a result of referring to S2, all rows that
have the same Department_code are retrieved, and therefore the overlapping is removed with
DISTINCT. Therefore, Department_code of departments that have less than five employees
that have Job_duty of Programmer are searched for, and Therefore a) is the appropriate
answer.
b) With the same method as a), the number of rows where Job_duty is not Programmer is
tabulated, and they are retrieved when the number of rows is over five. Having six or more
employees who are not a programmer does not necessarily mean that there are less than five
programmers.
c) According to the conditional expressions in the GROUP BY clause and the HAVING COUNT
clause, the table S1 is not narrowed down by job duty, and departments that have five or less
and even one programmer are searched for under those conditions.
d) Because the IN clause is used, S2 is referred to for one row of S1, and Department_code
where Job_duty is programmer is retrieved, and in that if Department_code is the same as S1
then Department_code of S1 is retrieved. Therefore, departments that have no programmer
are not searched for.

69
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-25 d) Deadlock when a table is updated

Deadlock means when multiple transactions attempt to lock a shared resource, they mutually
wait for the unlocking (releasing) of the resource locked by the other party, and they end up
perpetually waiting. Key points in this question are: locking is done for each table just before
update, and unlocking is done after the transaction is completed.
After the transactions start, Transaction A and Transaction B lock Table T1 and Table T2
respectively, and performs the update process ((3) and (4)). However, unlocking is done after the
transaction is completed, and therefore the process continues with tables remaining locked. Prior
to process (5), Transaction A attempts to lock Table T2. However, it is not able to lock as Table T2
is being used by Transaction B, and Transaction A gets into the waiting mode (At this stage,
Transaction A is in waiting mode, but deadlock hasn’t occurred). Next, prior to process (6),
Transaction B attempts to lock Table T1. However, it is not able to lock as Table T1 is being used by
Transaction A, and Transaction B gets into the waiting mode. At this stage (6), both Transaction A
and Transaction B are in mutually waiting mode for resources, and deadlock occurs. Therefore, d)
is the correct answer.

Transaction A Transaction B

(1) Transaction starts (2) Transaction starts

Table T1 locked Table T1

(3) Update Table T1 Table T2
Table T2 locked
Time

(4) Update Table T2

(5) Table T2 locked Wait
(6) Table T1 locked Wait

70
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-26 a) Exclusive control of database

When a database record is being used by a transaction, exclusive control of a database is a

function that restricts access to this database record by another transaction. In this exclusive
control (lock), there is a “shared lock” and an “exclusive lock.” For example, when a record is
used for reference purposes, it causes inconvenience if another transaction updates this record
while it is being referred to. However, it does not cause any inconvenience even if it is
simultaneously referenced by another transaction. In this case, a shared lock is used. In other
words, although reference by multiple transactions at a time is permitted, update is not allowed.
On the other hand, when a record is used for update purposes, it causes inconvenience if another
transaction refers to this record during the process of update. An exclusive lock is used in such
case. In an exclusive lock, the corresponding record is exclusively occupied, and even reference to
the record is not allowed. Therefore, a) is the appropriate answer. Although shared locks may
coexist, other combinations cannot coexist.

Q3-27 d) Files for database recovery

A backup file is a file where the entire database of a certain point in time is saved. In the case
of a media failure, once the media is recovered, by loading the contents of the latest backup file, it
is possible to return to the point where the backup was taken. However, it does not return to the
state when the failure occurred (down point). Log file (journal file) contains update history
(information before update and information after update) for every data item, and therefore the
information is recorded in chronological order. With this, the recovered database is updated to the
point right before the down point. At that time, information after update is used. This process is
called a rollforward process. Therefore, d) is the correct answer.
a) Transaction file: This file contains update information (daily transaction information) of the
master file.
b) Master file: This file is maintained as a base file. Like a journal, this file contains the most
basic information of operations.
c) Rollback file: This file contains information before update for rollback. However, generally it
is included in d) log file.

71
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-28 d) Techniques of recovering transactions

For a database recovery process, a log is stored in the database management system. In this log,
starting point of transaction, information before the data is updated, and information after update
are recorded. Moreover, by setting the checkpoint and by generating the log, a transaction
identifier (information for identifying the transaction) under execution is generated, and then
changes made up to that point in time are written to the database. As the checkpoint is obtained
when Transaction T is running, changes made up to obtaining the checkpoint are written on the
database. Therefore, in order to restore the database to the point of right after the completion of
Transaction T, it is necessary to re-run the process from the checkpoint recorded in the redo log
until the completion of transaction, and then write the result to the database. Therefore, d) is the
correct answer.
a) A two phase locking (2PL) is divided into two phases, namely a phase where locks are
acquired (growing phase) and a phase where locks are removed (shrinking phase). In the case
that a transaction needs multiple locks, instead of locking and unlocking each time to maintain
consistency, after the lock is applied, it is necessary to keep the lock until the writing is
completed.
b) Transaction scheduling decides the execution sequence of transactions and lock granularity so
that deadlock or inconsistency of update does not occur. Two phase locking in (a) is also a
type of transaction scheduling.
c) Roll back is a recovery technique for the restoration of the database to the state prior to the
transaction by using the undo log.

Q3-29 b) Data mining

Data mining is a technique of uncovering semantic information such as underlying rules and
relations in a large amount of historical data. For this, advanced mathematical techniques such as
artificial intelligence and statistics are used, and it attempts to predict the future based on past data.
As supporting tools for this, software programs equipped with a fast search function through a
large amount of data and automatic rule detection algorithms are developed. Therefore, b) is the
appropriate description.
a) Although data mining is used as one of the fast search methods, it is not limited only to that.
c) This describes data warehouse.
d) This describes data mart which is a part (or subset) of data warehouse and is rearranged for
each department.

72
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-30 a) Transparency of a distributed database

Distributed database refers to distributing and keeping data in multiple locations, and it
functions by mutual cooperation over communication lines. An application program that uses this
database can use the entire distributed database as a single database without realizing the physical
location of the data. Moreover, transparency of a distributed database means that data can be
handled without knowing the storage location of data or other circumstances specific to the
distributed database.
Databases reside on “multiple” servers and can be accessed “as if they are operating on one
server.” This means that a) is the appropriate choice.
b) An application program “needs not know” which server’s database should be accessed.
c), d) A distributed database cannot exist “on one server.”

Q3-31 b) Communication load between client and database server

An effective method for solving the problem of the communication load of SQL statements is to
use a stored procedure function that involves storing typical processes of frequently used SQL
statements in the database management system beforehand. Communication load between client
and server can be reduced with this function. Therefore, b) is the correct answer.
a) Redefinition of the index: Although it streamlines the database search, it does not result in the
reduction of communication load of the SQL statements.
c) Reorganization of the database: This is used when repetitive addition, deletion, and update of
data lower the storage efficiency of a database. It is not a solution for reducing the
communication load.
d) Use of dynamic SQL: Dynamic SQL is one of the methods of embedded SQL. As SQL
statements are still sent in this method, it is not an appropriate solution for reducing the
communication load.

Q3-32 a) Data buffering

As encoding speed and communication speed are represented in bits while size of audio data is
represented in bytes, the conversion of everything in units of bytes gives the encoding speed of 8
Kbps and a communication speed of 6 Kbps.
The time required for encoding 1.2 Mbytes of audio data is 1.2 Mbytes ÷ 8 Kbps = 150 seconds.
On the other hand, the time required for downloading is 1.2 Mbytes ÷ 6 Kbps = 200 seconds.
Therefore, it is clear that the downloading time is 50 seconds longer than the encoding time. For
playing back uninterruptedly, it is necessary to simultaneously complete downloading and
encoding. For simultaneously completing both of these, 50 seconds of data buffering is required
before the playback is started.
Therefore, a) is the correct answer.

73
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-33 a) Calculation of bit error

A bit error ratio of 1/600,000 means that an error of one bit occurs in 600,000 bits. A
transmission speed of 2,400 bps means that 2,400 bits are transmitted in one second. Therefore, the
number of seconds required for transmitting 600,000 bits should be calculated. The required time
is 600,000 (bits) ÷ 2,400 (bits/second) = 250 (seconds). Therefore, a) is the correct answer.

Q3-34 c) Role of DNS server in TCP/IP network

IP addresses used on the Internet at present are 32-bit binary numbers (for practical convenience
they are shown by dividing them into four sections of 3-digit decimal number each) that are
difficult for ordinary people to understand. Therefore, addresses using character strings called
domain names are used. The mechanism for searching for an IP address from a domain name is
called a DNS (Domain Name System). For example, it searches for the IP address from the domain
name www.abcdef.co.jp. A DNS server is a server that has the function of a DNS, and therefore c)
is the correct description.
a) This describes the directory service that manages network resources.
b) This is an explanation of NAT (Network Address Translation) and IP masquerade. Both are
methods for effectively using the limited number of available global IP addresses.
d) A DHCP (Dynamic Host Configuration Protocol) server has this function. Instead of
allocating an IP address to the relevant computer in advance, a DHCP server automatically
sets the IP address each time the client’s computer requests it.

Q3-35 d) NAT function of router used for Internet connection

By using the function that converts a private IP address to a global IP address and vice versa as
described in d), a device that has a private IP address can access an external network (the Internet).
The function that converts addresses in this manner is called NAT (Network Address Translation).
Although NAT is divided into narrowly-defined NAT that assigns a global IP address to each
device for accessing the Internet, and NAPT (Network Address Port Translation) that allows
multiple devices to access the Internet with one global IP address, NAPT is also included in
broadly-defined NAT.
a) This is an explanation of the caching function of a proxy server.
b) NAT is an address conversion function, and not a function that detects a bit pattern. In
addition, a virus check performed by antivirus software is one of the typical examples of
functions that detect specific bit patterns.
c) This is the explanation of a firewall.

74
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-36 a) 10-Mbps LAN based on CSMA/CD

The access control method in CSMA/CD (Carrier Sense Multiple Access with Collision
Detection) detects that the carrier is not passing through the transmission path, and then each node
starts sending data frames. At this time, if multiple nodes simultaneously send frames, data may
collide. The node that detects the occurrence of collision immediately stops sending frames and
sends out a jam signal. After a random period of waiting (called back-off time), it is checked
whether the carrier is passing through the transmission path, and if the carrier is not passing
through, transmission of frames becomes possible. Therefore, a) is correct.
b) In the case of CSMA/CD, time-division multiplexing of the transmission path is not
performed. When frames can be sent, they are sent at the rate of 10 Mbps.
c) The access control method that involves obtaining tokens is a token passing method.
d) Transmission method where timeslot is defined for each device, and then data is sent
according to the time slots defined is TDMA (Time Division Multiple Access).

Q3-37 c) Device for interconnecting LANs at the physical layer

When LANs are interconnected, there are predefined devices according to the connection
function. For interconnecting LANs at the physical layer (Layer 1) of the OSI basic reference
model, c) Repeater is used.
a) Gateway – A device that supports all protocol conversion functions from Layer 3 and above
up to Layer 7.
b) Bridge – A device that establishes an interconnection at the data link layer (Layer 2).
d) Router – A device that establishes an interconnection at the network layer (Layer 3).
Nowadays, a hub is widely used for connecting a computer to a LAN, and for interconnecting
LANs. However, the hub here indicates a repeater hub equipped with an amplification function at
the signal level just like a repeater. In addition to this, like a bridge, there is a switching hub that is
equipped with a filtering function based on a MAC address.

Q3-38 c) Description concerning inter-LAN connection devices

By connecting two LANs, it is possible to create another LAN that has even wider coverage. In
this case, the required devices differ depending on the function level of the layer of the OSI basic
reference model where the connection is established.
A repeater establishes a connection between segments at the physical layer. The signal is
amplified at the bit level, and no specific data identification is done. It merely extends the
transmission path. Therefore, c) is the appropriate answer.

75
 Morning Exam Section 3 Technology Elements Answers and Explanations

a) A gateway establishes an interconnection between networks that have different protocols in

Layer 4 through Layer 7.
b) A bridge uses the address (MAC number; Media Access Control number) used by the data
link layer (Layer 2) to determine whether that packet should be allowed to pass or not
(filtering), and enables regenerative relaying.
d) A router sets the relaying path for data (routing) in Layer 3 based on the IP address.

Q3-39 a) Specifications newly added or changed in IPv6

An IP address of IPv4 (Internet Protocol version 4) is 32 bits long. Growth in the number of
Internet users resulted in the problem of an insufficient number of IP address, and therefore IPv6 is
standardized as the next version of IPv4. In this IPv6 (Internet Protocol version 6), the bit size of
an IP address is changed from the 32 bits of IPv4 to 128 bits, and therefore a) is the correct answer.
In addition, IPv6 is further expanded so that automatic acquisition of IP address, security setting
functions for authentication and encryption, and specifying the path at the starting point can be
used as standard functions.

Q3-40 a) Obtaining MAC address from IP address

A protocol for obtaining MAC addresses from IP addresses is the ARP (Address Resolution
Protocol) stated in a).
In ARP, the host that wants to get a MAC address sends an ARP request packets containing its
known IP address to be inquired to the entire LAN. Every host that receives this ARP request
determines whether it is meant for its own IP address, and the host where the APR request matches
with the host’s IP address returns the ARP response packet containing the MAC address. The MAC
address obtained by APR is cached for some time; however, old information is automatically
deleted. In addition, the protocol used for obtaining an IP address from its known MAC address is
RARP (Reverse ARP).
b) DHCP (Dynamic Host Configuration Protocol) is a protocol for automatically allocating an IP
address when the client is started.
c) ICMP (Internet Control Message Protocol) is a protocol for determining and providing
notification of failures such as an IP packet not reaching its destination.
d) NAT (Network Address Translation) is a technique for one-to-one inter-conversion of private
IP addresses and global IP addresses.

76
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-41 b) Calculation of network address

In an IP address, the section of digit 1 in the subnet mask is the network part, and the section of
digit 0 is the host part. Writing the subnet mask of “255.255.255.240” in binary number gives the
following:

IP address 11001000 101 01010 01000110 00010011

Subnet mask 11111111 111 11111 11111111 11110 000
Network address 11001000 101 01010 01000110 00010000

An AND operation on an IP address and a subnet mask gives a network address, which in this
case is 200.170.70.16. Therefore b) is the correct answer. In addition, in the case of this subnet
mask, only the last eight bits are required to be calculated.

Q3-42 c) Address class of IP address

This question is about IP addresses.

By looking at the bit pattern of the first three bits in an IP address, it is possible to identify the
class. An IP address where the first three bits are 000 is Class A, 100 is Class B, 110 is Class C,
and 111 is Class D. Class D IP addresses are also called multicast addresses, and they are specially
used for simultaneous communication between specific participants only.
Based on the aforementioned bit patterns, the first three bits of the IP address starting with 192
are 110, which is Class C. Therefore, c) is the correct answer. Moreover, in this question, in the list
of IP addresses, a private IP address is also provided as an example.
a) In a Class A IP address, the first three bits must be 000. Therefore, this does not match with
192.168.10.10, where the first three bits are 110.
b) In a Class B IP address, the first three bits must be 100. Therefore, this does not match with
192.160.10.10, where the first three bits are 110.
d) In a Class D IP address, the first three bits must be 111. Therefore, this does not match with
192.160.10.10, where the first three bits are 110.

Q3-43 a) Protocol that dynamically assigns IP address

The protocol that efficiently manages IP addresses by dynamically allocating IP address in a

TCP/IP network is a) DHCP (Dynamic Host Configuration Protocol).
DHCP dynamically assigns IP address to the relevant clients when the client is started or when
there is a request for the assignment of an IP address, and recovers the assigned IP address when it
is shutdown. By dynamically setting network parameters in this way, even if the computer is
moved to another sub-network, it can be used without taking the settings of network parameters
into consideration.

77
 Morning Exam Section 3 Technology Elements Answers and Explanations

b) HTTP (HyperText Transfer Protocol) – Protocol for exchanging files, such as HTML files,
between a Web server and a Web browser
c) LDAP (Lightweight Directory Access Protocol) – Protocol using the directory service that
manages information about the environment and e-mail addresses of Internet users
d) SNMP (Simple Network Management Protocol) – Protocol that manages a network and
defines the method for data exchange between the management system and network devices
under the scope of management.

Q3-44 d) Upper-layer protocol of IP

As IP falls under the network layer of the OSI basic reference model, the transport layer above it
should be considered. The transport layer achieves communication between applications
(processes) residing on different hosts. Although the transport layer has protocols such as TCP and
UDP, the connectionless protocol where flow control, sequence control, and establishment of
connection are not performed is UDP (User Datagram Protocol) in d).
Although the reliability of UDP is low, it is easy to implement and offers high speed service, and
therefore it is used for real-time applications of video and audio such as Internet broadcast and
Internet telephony, and applications that exchange a small amount of data intermittently.
a) ICMP (Internet Control Message Protocol) – In the case of an error in a router or a host, this
protocol provides notification of the error status to the source. It works in the network layer of
the OSI basic reference model.
b) PPP (Point to Point Protocol) – This protocol located below IP connects two sites over a WAN
in a multi-vendor multi-protocol environment, and works in the data link layer of the OSI
basic reference model.
c) TCP (Transmission Control Protocol) – This is a highly reliable protocol for an application
that continuously handles large amounts of data. Like UDP, it works in the transport layer of
the OSI basic reference model; however, it establishes a physical connection and has the
functions for acknowledgement and sequence control.

Q3-45 d) Network management protocol of TCP/IP

The network management protocol used in the TCP/IP environment is d) SNMP (Simple
Network Management Protocol). Information about the failure of devices connected to the network
and about the connection can be tracked, and the load on each device can be monitored based on
the amount of data sent and received.
a) NNTP (Network News Transfer Protocol) – This protocol is used for reading and posting
articles on the Netnews where in-group discussions are held on the Internet, and groups are
formed for each of the specific topics.
b) NTP (Network Time Protocol) – This protocol is used to adjust the internal clock of a
computer by accessing a server located on the network and obtaining the correct time.

78
 Morning Exam Section 3 Technology Elements Answers and Explanations

c) SMTP (Simple Mail Transfer Protocol) – This protocol is used for the exchange of e-mails
between mail servers. Moreover, e-mails are forwarded from the client to the server by using
SMTP. However, for delivering e-mails to the client from the mail server, other protocols such
as POP and IMAP are used.

Q3-46 a) Technology enabling virtual dedicated network

VPN (Virtual Private Network) is a technology that uses authentication technology and
encryption technology so that a public line can be used like a virtual dedicated line. Originally, it
was technology for using telephone lines, but as Internet connections became commonly available
IP-VPN was designed for usage on the Internet, which is a public network. Its advantage is the
cost reduction compared to actually building a dedicated line. Therefore, a) is the correct answer.
b) IPv6 is the latest standard of the IP protocol used on the Internet. The IP protocol widely used
at present is IP version 4 (IPv4), and it is composed of 32 bits. However, due to rapid growth
in the number of Internet users, there were concerns that IP addresses would be exhausted.
Therefore, IP version 6 (IPv6) was formulated for solving this problem. Since IPv6 is
composed of 128 bits, it allows the use of an enormously large number of IP addresses in
comparison with IPv4. In addition, packet encryption and authentication of packet sender are
provided as standard functions.
c) PBX (Private Branch Exchange) is a private branch exchange used in companies. It is used for
the connection between internal phones and between public line and internal phones.
d) VoIP (Voice over Internet Protocol) is a technology that implements the so called “Internet
phone,” where audio is converted into digital data and stored in IP packets for
communication. Special tools are required for using it, and in many cases, the counterparties
that can be dialed are also restricted. However, since a normal IP network is used,
communication charges are less expensive and independent of talking distance.

Q3-47 b) Session management in a system that is using browser

The HTTP protocol basically releases a session in a one-time data transaction. Therefore, with
the information specified, it is difficult to quickly change the contents of the next webpage. In
order to solve this, the Web server transfers the data for user identification to the browser, and by
presenting this data to the Web server when the browser accesses the Web server again, the Web
server identifies that access is from the same user and can provide a series of services. The
mechanism that enables this process is b) Cookie.
a) CGI (Common Gateway Interface) is an interface that runs a program (CGI program)
prepared on the Web server based on the browser’s request, and transfers the results to the
browser. While a normal webpage is a static file, dynamically created webpages can be
delivered with CGI.
c) SSL (Secure Sockets Layer) provides safe communication that is encrypted. It allows the
exchange of private or money-related information on the Internet without allowing another
person to access this information.

79
 Morning Exam Section 3 Technology Elements Answers and Explanations

d) URL (Uniform Resource Locator) is a description method for indicating the location of
information resources on the Internet, and it consists of protocol name, domain name, and file
path.

Q3-48 c) Explanation of SMTP

SMTP (Simple Mail Transfer Protocol) is a protocol for forwarding e-mails on the Internet. It is
used when e-mails are forwarded between mail servers and when a user sends an e-mail to the mail
server. Moreover, when the user receives e-mails from the mail server, the protocol known as POP
is used. Therefore, c) is the correct answer.
a) This describes HTTP (Hyper Text Transfer Protocol). When a URL which is a Web site
address is specified, the “http://” at the beginning of the URL indicates this protocol.
b) This describes DTP (Desk Top Publishing). DTP is the task of creating and editing block copy
of publication matter on a computer or a workstation, and the printing it with a printer or an
image setter. Data to be handled can also be a combination of characters, diagrams, and
photographs.
d) This describes a document description language also known as a markup language. Using
commands called tags enclosed with < >, the document structure and layout other than
characters can also be specified. Some markup languages are as follows:
HTML (Hyper Text Markup Language)
SGML (Standard Generalized Markup Language)
XML (eXtensible Markup Language)

Q3-49 b) Explanation of ADSL splitter

A splitter is a device that separates and combines an audio signal and data signal in a high speed
digital communication service such as ADSL that uses subscriber phone lines (copper lines). It is
installed at the both ends of a telephone line (switching center and telephone subscriber). In
ADSL, as audio signals and data signals pass through the same line, noise is mixed and this slows
down the transfer rate. Therefore, by using a splitter, incoming signals are separated into signals
for telephone equipment and signals for an ADSL modem. On the other hand, a splitter combines
outgoing signals and passes them through the telephone lines. Therefore, the answer is b).
a) A splitter is connected to a telephone line.
c), d) As described earlier, the only function of splitter is the separation and combination of
signals. It is not the ADSL transmission device, and it does not have any error correction
functions.

80
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-50 c) Combination of keys in public key cryptography

Public key cryptography forms a pair of keys: a public key and a private key, and shares public
keys with multiple senders. The private key is kept confidential by the recipient. While data is
encrypted with the public key, the encrypted data is decoded with the private key.
In this question, when the shop receives orders over a network, customers use a public key, and
the shop uses the private key. This method is safe because even if wiretapping is performed by a
third party, only the party (shop) that has the private key can decode the data. Therefore, c) is the
correct combination.
Public key cryptography

Customer Encrypted Encrypted data Shop

(Sender) (Recipient)
data

Forward
Message
(Plain text)
+
Encryption Decoding +

Message
(plain text)
Public key of
sender (Key pair)
Private key of
receiver

Typical public key cryptography: RSA (Rivest Shamir Adleman)

Q3-51 a) Purposes of using digital signature for communication

In a digital signature, the sender attaches a signature created by signing with plain text (refers to
unencrypted data), which allows the recipient verify the sender. A signature is an encrypted
message digest created from the plain text, and therefore, if the plain text differs, the message
digest also differs.
There are two purposes of using a digital signature.
- Personal authentication function: The sender encrypts the data with his or her own private
key (signing key), and if the recipient can decode it with the public key of the sender, the
recipient can verify the sender. Only the sender, who creates the pair of public key and
private key, can encrypt so that the encrypted data can be decoded with the public key of the
sender.

81
 Morning Exam Section 3 Technology Elements Answers and Explanations

- Message authentication function: If the message is altered, the message digest created from it
is different from the signed message digest. Therefore, it can be checked whether the
message has been altered or not after the signature is added (for the purpose of preventing the
message from falsification).
Therefore, a) is the correct answer.
b) From the mechanism of the Internet, neither wiretapping of the message itself can be detected,
nor can it be prevented. Therefore, it is not possible to check whether someone else read the
message or not.
c) Although a digital signature’s objective is user authentication, it is not used for checking the
ID of the sender.
d) As the sender is required to confidentially preserve the private key, it is not possible for the
recipient to check whether to return it or not.

Q3-52 d) Handling keys in public key cryptography

In public key cryptography, the encryption key and the decryption key are different, and data is
kept confidential only with the decryption key. Thus, it is common to publish the encryption
algorithm. Therefore, d) is the appropriate description.
Moreover, as public key cryptography ensures that data is safe from code-breaking, the key is
large at about 500 bits, and therefore the encryption and decryption speeds are much slower
compared with secret key cryptography. However, it has the advantage that when only one person
is required to receive the communication among n number of unspecified users, this can be easily
achieved with one decryption key (private key).

Q3-53 d) Protocol for enhancing the confidentiality of e-mails

Protocols for enhancing the confidentiality of e-mail content by encrypting the e-mails before
they are sent include PGP (Pretty Good Privacy) and S/MIME (Secure Multipurpose Internet Mail
Extensions). Therefore, d) is the correct answer. However, just like normal e-mails, SMTP and
POP are used for sending and receiving e-mails that are encrypted with PGP or S/MIME.
The meanings of the other terms are as follows:
a) IMAP4 (Internet Message Access Protocol version 4) – This is a protocol that is used for
receiving e-mails stored in the mailbox of the mail server.
b) POP3 (Post Office Protocol version 3) – Like IMAP4, it is a protocol for receiving e-mails
stored in the mailbox of the mail server. In POP and IMAP, the method of management for the
received e-mails is different. In other words, in POP, all e-mails are downloaded to the client,
and therefore e-mails are managed by the client. However, in IMAP, e-mails are not
downloaded on the client; rather they can be managed by the server.
c) SMTP (Simple Mail Transfer Protocol) – The protocol for sending e-mails from the mail
client to the mail server, and exchanging e-mails between mail servers in the TCP/IP network.

82
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-54 a) Purpose of using message digest

This question concerns a message digest.

The concept of a message digest is slightly different from what is generally called a digest. A
completely different digest value is generated even if one bit is different, and it is used for
checking in the case of falsification. Therefore, the description of a) is about a message digest.
MD5 and SHA-1 are two famous algorithms for this, and in MD5, a digest value of 16 octets is
generated, while in SHA-1, a digest value of 20 octets is generated.
As a message digest is widely used in combination with encryption technology for secured
communication, it is often confused as an encryption technology. However, it should be
remembered that “encryption is for hiding” and “message digest is for detecting falsification.”
b) This describes negotiation. Prior to the communication stage, adjusting both ends for
encryption technology and encoding method is called negotiation. In SSL (Secure Sockets
Layer), negotiation of encryption specifications is performed, while in IP phones, negotiation
of the CODEC is performed before actual communication.
c) This describes what is commonly known as a digest.
d) This describes encryption, and it is not related to the prevention of falsification.

Q3-55 c) Authentication between two communication entities

In Procedure (1), Y does not know whether the receiver is X itself or not.
In Procedure (2), X does not know whether the character string sent is from Y or not. However,
X generates a new character string based on the rule decided with Y beforehand, and returns it to Y.
Y already knows the rules that enable Y to read this newly generated character string.
In Procedure (3), Y decodes the returned character string based on the rule decided with X
beforehand, and checks that the contents are correct. If the contents are correct, it means that X is
the genuine other party with whom the rules were decided beforehand and therefore Y can
authenticate X.
In this case, it is important that the “rule decided between X and Y” is not known to a third party.
Therefore, X usually has a private key available only to X, and Y has a key (public key, or private
key) that can decode information that is encrypted by X.
In this procedure, X merely receives “character strings including random information” from Y,
and does not receive any other information about Y. In other words, X cannot authenticate Y.
Therefore, “c) Y authenticates X” is the correct answer.

83
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-56 d) Authentication methods for remote access

Internal systems can be remotely accessed from outside with different methods such as using
public lines like analog telephones and ISDN, and accessing over the Internet. In either of these
methods, there is a threat of unauthorized access by spoofing. Therefore, authentication is
performed to check that genuine user is accessing the system.
The most basic authentication method is authentication with user ID and password. User ID and
password are registered in an internal system beforehand, and authentication is performed at the
time of remote access. In the usual authentication that is based on fixed passwords, leakage or theft
of a password may result in unauthorized access by spoofing using this password. Therefore, as a
measure for strengthening the password authentication method, a one-time password method
where password can be used only once, was designed. As the authentication information
circulating on the channel is of the use-and-throw type, it cannot be reused even if it is intercepted.
For generating authentication information, there is the challenge response method that uses a
random number, and there is the synchronization method that synchronizes counter data, which
creates authentication information. A one-time password can be effectively used on public lines as
well as the Internet. Therefore, d) is the correct answer.
a) Registration of the IP address of permitted PCs in advance, and allowing communication from
only registered addresses while denying any communication from other devices is effective
for the enhancement of security. However, it does not result in enhanced authentication of PC
users if their PCs are fraudulently acquired.
b) In call back, during remote access over a public line, an internal system disconnects the
communication at once, and then reconnects by making a call to the number of the remote
device registered in the server. Although it strengthens communication, it is not applicable to
remote access over the Internet.
c) This does not result in enhanced authentication because there is a possibility of obtaining the
user ID by illegitimate means that lead to unauthorized access by spoofing.

Q3-57 b) Characteristic of security protocol SSL

SSL (Secure Sockets Layer) is a security protocol developed by Netscape Communications

Corporation of the United States for ensuring safe HTML communication. Nowadays it is
implemented in major Web browsers and Web servers as a standard function, and SSL is easily
available for use. In a Web server using SSL, by embedding the digital certificate in its own FQDN
(Fully Qualified Domain Name, complete URL), it can be checked to determine if it matches the
URL requested by the Web browser. Therefore, b) is the correct answer.
a) In TCP/IP, SSL exists in the application layer, while in the OSI basic reference model it exists
in the session layer. In addition to Web servers, it is also implemented in FTP and TELNET.
c) The certificate used in SSL can be stored in the file format by exporting it in a Web browser. For
example, in Internet Explorer, this can be done by exporting the certificate in the Contents tab under
the Internet Options menu. Moreover, by importing this file, it can also be used in other computers.
d) SSL is an industry-standard security protocol on a global scale, and there is no restriction that
128 bits can be used only by governmental institutions.

84
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-58 d) Act corresponding to social engineering

Social engineering refers to the act of using physical means to obtain the required ID and
password for unauthorized access to network systems. It also includes the acts of impersonation
and obtaining important information about a company or an organization from paper waste.
Therefore, d) is the correct answer.
a) Examples of types of attacks on a security hole in an OS (security flaws and vulnerabilities
arising from software design error) include DoS attack (attacking the devices constituting the
network, for interrupting the provision of services, and it is also known as denial of service
attack and suspension of service attack) for system shutdown and taking control of privileges
by command injection (Administrative rights are obtained by entering from the security hole
of an OS. All resources can be accessed).
b) The creation of a virus has nothing to do with social engineering.
c) In a dictionary attack, character strings listed in a dictionary are encrypted with the same
method as that used when a password file is created. After the dictionary entry which fully
matches with these results is found, the original password is identified by looking for items
that match these results.

Q3-59 d) Phishing

Phishing is a fraudulent act of illegitimately obtaining personal information by using a fake

e-mail that appears to be that of a real financial institution or online shopping Web site. With the
URL noted in an e-mail sent to multiple unspecified recipients, recipients are tricked into visiting a
fake Web site that looks exactly like the Web site of the actual company. If the user does not notice
that the Web site is fake and enters his or her personal information such as login ID, password,
name, address, and credit card number, this information is sent to the phishing source. Therefore,
d) is the correct answer.
a) This refers to cross-site scripting. Script not allowed by a user is automatically executed,
resulting in the risk of leakage of cookies.
b) This is unauthorized access through a backdoor. The backdoor itself is a window for
unauthorized access, and it may be automatically set up due to virus information, or it may be
created through manual operation by a cracker (an intruder).
c) This refers to spyware. It is software that runs in the background without the user’s
knowledge, and collects personal information and the computer usage pattern of a user and
then sends them to a marketing company or the person who developed the spyware.
Generally, it is embedded when some kind of application software is installed. Usually, the
user’s permission for installation is obtained at that point in time, so it is not necessarily
illegal. However, as there are few users who carefully read terms and conditions, this itself is
regarded as a problem.

85
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-60 b) Pattern matching method of antivirus software

Pattern matching methods of antivirus software compare and check the program with known
viruses to detect the presence of a virus. A signature code refers to the fingerprint of a computer
virus, and it is a unique code that exists in specific computer virus. Therefore, antivirus software
can identify and detect viruses from these signature codes. Therefore, b) is the appropriate answer.
a) A virus cannot be detected just by a comparison of files before and after infection to check if
any changes were made.
c) A method for the detection of a virus by monitoring abnormal phenomena attributable to the virus is
not a pattern matching method. Moreover, antivirus measures fall behind the curve in this method.
d) Matching with the checksum of a file is one of the methods known as generic detection, and it
is used as a means for the detection of suspicious programs and unknown viruses. Therefore,
it is different from the pattern matching that can specifically identify known viruses, and it is
not generally used as antivirus software like option b).

Q3-61 d) Role of risk analysis in security measures

Risk means “possibility of occurrence of event that interferes with the performance of corporate
activities and the results of such interference.” In risk analysis, the possibility of occurrence of a
security failure and its impact on information assets is evaluated. In other words, the extent of the
loss is obtained as an outcome of risk analysis along with the frequency of occurrence. Therefore,
d) is the correct answer.
a) Vulnerabilities are obtained as an outcome of the security process.
b) Implemented security controls are obtained as an outcome during the implementation of
security measures.
c) Security specifications are obtained as an outcome of the formulation of a plan for security measures.

Q3-62 d) Steps of implementing security management system of JIS Q27001:2006

JIS Q 27001:2006 describes the requirements to be satisfied when information security

management system (ISMS) is implemented in the organization. In this, it is mentioned that to
build the system, an organization should follow the six main steps described below.
(1) Define the ISMS policy of the organization.
(2) Define the scope of ISMS.
(Note: Scope is a concept that includes applicable range and objectives)
(3) Risk assessment (evaluation)
(4) Risk management
(Note: Management here refers to the act of identifying potential risks that may arise and their
resolutions)
(5) Select the controls to be incorporated and describe the target of these controls in detail.
(Note: Controls indicate methods and means of regulation)
(6) Prepare a Statement of Applicability (SOA).
Option d) describes this sequence, and therefore d) is the correct answer.

86
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-63 b) Information security policy in ISMS conformity assessment scheme

In the ISMS certification standards (Ver. 2.0) used in the ISMS conformity assessment scheme,
under the basic policy of information security in the annex “Detailed Controls,” the control
objective described is “To provide management direction and support for information security.”
This basic policy of information security is also called information security policy, and therefore
b) is the appropriate description.
a) Information security measures should be implemented company side, and its basic policy
(security policy) should be informed everyone in the organization.
c) ISMS should be built, implemented, and maintained as a management system, and it should
be continuously improved. Basic policy (security policy) should also be periodically
reviewed, and it should be continuously improved for optimizing its contents in accordance
with changes in business environment and technologies.
d) Information security policy sets the overall direction and principle concerning information
security in the organization, and it is not a detailed description of a particular system.

Q3-64 d) Response procedure in the event of occurrence of security failure

In case of detection of unauthorized access to a Web server from an external location, in order to
prevent the damage from spreading, it is necessary to promptly disconnect the Web server from the
network. A description of this is provided in (3).
Next, a survey should be conducted to determine the extent of damage. In concrete terms, each
log of the server, IDS (Intrusion Detection System), and firewall should be analyzed, and the
intrusion path and method as well as the range of the unauthorized access should be identified.
This is noted in (1).
Next, the damaged area that was identified should be reinforced, and the network reconstructed.
Especially, the intrusion path is the area that does not have adequate security, so it is necessary to
take adequate measures henceforth. In concrete terms, the latest patches and security setting
information that are revised based on the damage should be applied. With regard to the Web server
also, falsified content and the damaged area should be restored. This is mentioned in (2).
As per the aforementioned procedure, the network should be reconstructed, and the Web server
that is now safe should be connected to the network. Since the intruder may attempt unauthorized
access again if security is not adequate, temporary monitoring should be strengthened and the
situation should be closely watched. This is noted in (4). Therefore, d) is the correct answer.

87
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-65 a) Human act of wrongfully obtaining the confidential information

Instead of using electronic methods, the act of illegitimately obtaining a password or other
confidential information using administrative deficiencies attributable to human behavior, or
deficiencies in physical security is called a) social engineering. For example, it includes acts such
as, feigning the identity of the user and making the request to the system administrator such as “I
forgot my password so please tell me it because I am working on something very urgent” to obtain
the password, or pretending to be the system administrator and asking the user to give his or her
password. In addition, shoulder surfing, which means covertly looking over someone’s should
when he or she is entering his or her password with a keyboard, and scavenging, which means
obtaining important information by recovering the discarded forms and other material from a dust
bin are also social engineering.
b) A Trojan horse is a form of computer virus. Although it may appear normal from the outside,
it is an unauthorized program that executes a malicious code that functions as a backdoor and
damages the system or leaks information. Like a worm, it does not self-diffuse, and it exists as
an independent program without infecting other programs.
c) Password crack refers to illegitimately analyzing a password with electronic methods such as
a dictionary attack or a brute force attack.
d) In the stepping-stone attack, instead of directly attacking the target, the attacker indirectly
attacks through a network by using another site that has vulnerabilities. It attempts to hide the
identity of the attacker.

Q3-66 d) Measures for preventing theft of registered password

A hash value is a value of fixed length determined from the data assigned by a certain algorithm.
The characteristics of the hash values are “If the original data differs, hash value derived from it
usually differs, and the hash value is rarely the same,” and “even if the hash value and the hash
function are known, the source data cannot be identified.” Therefore, if a password is converted
into a hash value, the original password cannot be recovered from the hash value even if the
password file is stolen, and hence d) is the correct answer.
a) Covering a user ID with the hash function has nothing to do with password theft prevention.
b), c) If the entire password file is stolen, fraudulent use of stolen passwords cannot be
prevented, and unauthorized users cannot be kept out.

88
 Morning Exam Section 3 Technology Elements Answers and Explanations

Q3-67 d) Management approach for preventing falsification and desctruction of data

In order to maintain the convenience for users whereby they can continue sharing a program,
setting access restrictions for data processed by the program is an effective management approach
for preventing falsification and destruction of the data. Therefore, “d) Setting of access rights for
files” is the appropriate preventive measure.
a) Taking an access log is certainly useful to ascertain whether unauthorized execution of a
program occurred based on access records, and also for the recovery of damaged data with the
update journal, but it cannot prevent damage to the data.
b) As the program itself may be altered and it may be executed illegitimately, “Comparison of
the source program and the executed program” is helpful in ascertaining the cause of damage,
but it cannot prevent damage.
c) Storing source programs in multiple locations certainly acts as an antitheft measure for the
source programs, but it cannot prevent damage.

Q3-68 d) Explanation of Web beacon

Web beacon refers to an image embedded in e-mails of HTML format or webpages to collect the
access information of users. Therefore, d) is the correct answer. Usually, it is a very small image
which is not noticed by users. If users access the Web site or e-mail where a Web beacon is
embedded, a server is accessed to fetch that image data. The server records this access
information, and secretly collects the access pattern of users.
Beacon actually means “indicator”. Since it is a very small image, it is also called a Web bug in
many cases. This Web beacon is the subject of heavy recent criticism, and some e-mail programs
can be set so that image files do not open during e-mail preview.

Q3-69 c) Connection restrictions in a wireless LAN

In a wireless LAN, MAC address filtering restricts access by terminals not registered at the
access point beforehand. A MAC address is a fixed physical address assigned to the NIC (Network
Interface Card) of the terminal. Therefore, by registering the MAC address of the permitted
terminal at the access point beforehand, the connection of unregistered devices can be restricted.
Therefore, c) is the correct answer.
a) AES (Advanced Encryption Standard) – This is an encryption algorithm that uses common
key cryptography.
b) IEEE 802.11b – It is a communication standard for a wireless LAN that uses the 2.4GHz
band. There is no access restriction for each terminal.
d) TKIP (Temporal Key Integrity Protocol) – It is a protocol that enables encrypted
communication and detects the falsification of frames in a wireless LAN.

89
 Morning Exam Section 4 Development Technology Answers and Explanations

Section 4 Development Technology

Q4-1 c) Software requirements

Software requirements (also called functional requirements) are for the fulfillment of business
requirements. In other words, software requirements are for the implementation of business
processes in a business system. The definition of data is an important software requirement.
Therefore, c) is the correct answer.
a) The disk capacity of the development environment is a necessary system requirement for system
operation. It is not a software requirement, and is defined as a non-functional requirement.
b) The objectives of computerization are specified at the planning phase of the system.
d) Database design is performed at the external design phase.

Q4-2 a) Identifying notation

The notation used in the question is a) DFD (Data Flow Diagram). A DFD primarily describes
functions (processes) and the flow of data among these functions. The symbols used are shown
below.

Process
Data store (file)
Source (origin) or sink (destination) of data
Flow of data among components (data flow)

b) A state transition diagram is a diagram for representing the events that cause changes in state,
and their accompanying changes in state, for objects that can take one of several different states
depending on the occurrence of events.
c) A flowchart uses flowchart symbols to represent the flow of data and control.
d) A Petri net represents the synchronization among functions that operate in parallel. The
generation of water through the reaction of oxygen and hydrogen is shown below as an example.

90
 Morning Exam Section 4 Development Technology Answers and Explanations

Before reaction After reaction

State (place) Hydrogen (H)
Hydrogen (H)
Event (Transition)
2
2
1 1
1 1
Water (H2O) Water (H2O)

Oxygen (O) Token Oxygen (O)

Fig. A Petri net model of the generation of water through the reaction of oxygen and
hydrogen

Q4-3 c) Code design for business systems

Appending a check digit to code allows the strengthening of code input checking. Therefore, c) is
an appropriate statement.
a) Actual assignment of code is generally performed by system users.
b) The attributes and number of digits in a code must be designed with consideration of not only
the internal processing efficiency of the computer, but also other factors, such as ease of use
from a user perspective, commonality, systematization, and expandability.
d) It is difficult to change the scheme of code once the code begins to be used. Therefore, it is
necessary to fully consider maintainability and expandability at the time of design.

Q4-4 a) UML standardized by OMG?

UML (Unified Modeling Language) is a modeling language that is based on an object-oriented

approach. It is used in software development and was standardized by a body called the OMG
(Object Management Group). Therefore, a) is the correct answer. In order to describe the details of
analysis and design, UML has various diagrams such as use case diagrams, class diagrams, sequence
diagrams, and state chart diagrams.
b) This is a description concerning IDL (Interface Definition Language).
c) This is a description concerning SQL (Structured Query Language).
d) This is a description concerning SGML (Standard Generalized Markup Language).

91
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-5 d) Target for modeling with use cases

As described in the question, a use case is “a technique that clearly describes the functional
requirements of a system and defines interactions between a user (actor) and the system.” In UML,
interactions between an actor and a system is modeled and expressed with use case diagrams. Use
cases, which are constituent elements of a use case diagram, are used to signify operations performed
by the system for providing services to the actor, and in a broad sense, use cases can be considered as
a unit of function that is provided by the system. Recently, as UML has gained popularity, use cases
have become common as constituent elements of a use case diagram. However, as described in the
question, use cases are also known as a method or technique for clearly specifying the functional
requirements of a system. In either case, use cases are used for the specification of the functional
requirements of a system, and if the option that is related to the function of a system is chosen, the
obvious choice is d). On the other hand, a) and b) are not implemented with a function of an
information system, and c) is a processing method. Therefore, none of these choices are appropriate.

Q4-6 d) Explanation of encapsulation in object-oriented approach

Encapsulation refers to unifying data and procedures related to it, and then concealing it inside an
object. By encapsulation, only operations (methods) performed on the object are shown externally.
Therefore, d) is the appropriate answer.
a) Abstraction of multiple objects with common characteristics is called generalization (the
opposite of which is specialization).
b) Making the subclass (child class) inherit the characteristics of a base class (parent class, super
class) is called inheritance.
c) “Abstracting common characteristics between classes and creating a base class” is a description
of generalization.

92
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-7 c) Method for transmitting information with the weakest module coupling

Module coupling is a measure of interdependence between modules, and it is categorized into six
types. Generally, the weaker the coupling, the better it is considered to be. Arranging in the sequence
from strong to weak gives the following result. “No direct coupling” (7) means that there is no
interdependence between modules, and it is the ideal form of relationship. However, this is very rare
in practice, and therefore coupling is usually categorized into the following six types from (1) to (6).
(1) Content coupling
(2) Common coupling
(3) External coupling
(4) Control coupling
(5) Stamp coupling
(6) Data coupling
(7) No direct coupling
Among these, in “data coupling” (6), which has the weakest module coupling, “only data items are
passed as arguments between modules,” so the correct answer is c). In “stamp coupling” (5), when
the pointer of a structure is passed as an argument, the receiving side will become aware of this data
structure and use the required data.
a) “Data defined in a common area is referenced by the relevant modules” corresponds to
“common coupling” (2).
b) “Control parameters are passed as arguments, and the execution sequence of modules is
controlled” corresponds to “control coupling” (4). In this form, awareness of the internal logic of
other modules is necessary, so module coupling strengthens.
d) “Required data is shared through external declaration” corresponds to “external coupling” (3).

93
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-8 c) Module having the highest module strength

Module strength is a measure of the strength of the connection between instructions in a module,
and the stronger the connection (high strength), the better it is considered to be. “Functional
strength” is required for a module to perform its unique function. In other words, all instructions are
required to perform this one function. As the module with the highest “functional strength” is
considered to be the best, c) is the correct answer.
As mentioned above, module strength is a measure of strength of the connection between
instructions in a module. The objective of considering this measure is to increase the maintainability
of a module. When one module has multiple functions, and if the connection between these functions
is not strong, maintenance of the module becomes difficult (for example, the modification of one
instruction may affect multiple functions). Therefore, modules that have a “single function”
(functional strength) with a high strength are considered to be better.
With regard to other options in the answer group, a) includes multiple functions for identical data,
so it is called “communicational strength.” As for d), it includes multiple functions performed
simultaneously at a specific point in time, and it is called “classical strength.” Moreover, with regard
to b), it is difficult to determine the strength from the description. However, data processing from
different media means that the module has multiple functions, and therefore it cannot be said that its
strength is higher than module c), which has a single function.
Incidentally, in addition to three types of strengths (functional strength, communicational
strength, and classical strength) described above, there are four other types of module strengths. Each
of them corresponds to the following modules.
・ Coincidental strength: The function of a module cannot be defined. A large program is simply
partitioned into modules with number of lines or another parameter as a guide.
・ Logical strength: A module has multiple functions, and a function is executed in a selective
manner according to the details of the function parameters.
・ Procedural strength: A module that contains multiple functions defined in functional
specifications.
・ Informational strength: A module into which multiple functions that handle the same data are
combined, and which has separate entry point names for each function (module name is separate
for each function).
These seven strengths can be arranged in the decreasing order of strength (strong, desirable) as
follows: functional strength, informational strength, communicational strength, procedural strength,
classical strength, logical strength, and coincidental strength. Therefore, module partitioning that
results in modules with functional strength and informational strength is desired.

94
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-9 a) Review method of a design document

In system development, work is divided and assigned to multiple people, and these people work in
parallel. If work is performed by the person in-charge only, assumptions or mistakes may cause
incorrect results or deliverables. In order to prevent this, upon completion of each process, the person
in-charge and other relevant persons get together and conduct a review to check the content of the
work. A series of activities conducted by relevant persons reviewing the document or work in order
to detect design errors as early as possible is called a) walkthrough. Furthermore, a review conducted
under the guidance of a dedicated facilitator (called a moderator) to identify and analyze problems
and consider measures for such problems is called an inspection.
b) Desk checking: A debugging method where the developer himself traces instructions one by one
with the assumed test data without running the program. Errors may be overlooked due to
perceived notions.
c) Top-down testing: A method for an integration test, where modules are integrated and tested in
sequence from the upper level to the lower level.
d) Parallel simulation: In a system audit conducted using a computer, an auditor prepares a test
program for the audit with respect to a specific application, and performs verifications by
comparing the simulation results obtained using the actual input data with the actual results
generated by the production program.

95
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-10 a) Situation that can be anticipated from bug control chart

As all three curves in the bug control chart given in the question remained at the same levels, the
chart indicates that all three activities of test execution (number of test items not executed), bug
detection (number of bugs detected), and bug resolution (number of unresolved bugs) have halted,
and the test is not progressing. There could be multiple reasons for this, such as all members of the
test team caught a cold or went on a company outing. However, if a selection is made from the
answer group, as mentioned in a), it can be thought that the test team have discovered a bug that is
difficult to resolve. For example, a functional bug that is related to all test cases not yet executed
(remaining) has occurred, and unless this bug is resolved, testing cannot move forward, which leads
to the situation described in the question.
In the typical form of the software reliability growth curve, the rate of bug detection is low
immediately after the test starts. This is usually because tests have come across a critical bug that
affects other tests. As is the case in this question, testing cannot move forward.
b) Rather than the detection of many bugs, the reason that the execution of test items is not
processed further could be that the number of unresolved bugs is not decreasing.
c) The number of test items not executed is flat but it is not approaching zero. Therefore, it cannot
be said that test is close to being completed.
d) The fact that the number of bugs detected becomes equal to that of test items executed has no
relation with the resolution of bugs.

Q4-11 b) Black box test

The method for the design of test cases is divided into a white box test and a black box test.
A black box test is a test to check whether the program satisfies the functions in the design
document without focusing on the internal logic of the program. It can also be said that this test is
conducted from the viewpoint of a user. Equivalence partitioning and boundary value analysis are
used for preparing test data. In this case, since focus is not on the internal logic of the program, even
if there is unnecessary description (redundant code) in the program, it cannot be detected unless
processing results are abnormal. Therefore, b) is the appropriate description.
On the other hand, a white box test is a test conducted based on the internal specifications of a
program, and test data is prepared in consideration of the coverage of instructions, conditional
decisions/branches, path combinations, and other factors. Description of a), c), and d) corresponds to
the white box test.

96
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-12 c) Combining test data of equivalence partitioning technique

This question is related to design methods for test data in a black box test. A black box test is a test
that checks whether functions are executed correctly based on the external specifications without
examining the internal specifications of a program.
For test data, equivalence partitioning, which is a design method for test data, uses the
representative values of a valid equivalence class (which is valid data) and the representative values
of an invalid equivalence class (which is invalid data). Here, since minimum test data is determined
by taking one representative value from the invalid equivalence class of – 2 to 0, the valid
equivalence class of 1 to 5, and the invalid equivalence class of 6 to 8. This means that a total of
three representative values are selected as test data.
Therefore, c) – 1 , 3, 6, which satisfies this, is the appropriate answer.

Q4-13 c) Verification test

In system modification during the maintenance phase, modifications that are made often adversely
affect other functioning sections, and functions that were working normally before modification stop
working correctly. Therefore, for the testing conducted after system modification, in addition to
testing the modified section, it is necessary to check that functions that were working normally
before the modification still work correctly. A test that verifies that such modifications do not lead to
new errors in functions that were working correctly is called c) regression test.
a) A function test is a test for verification that a system satisfies all the functions required by its
specifications.
b) An integration test is conducted after a unit test of each module is completed, and it verifies that
modules are processed correctly when they are integrated.
d) An exception test is a test for verification that when data other than the predefined data is
processed, such data is processed as per specifications without abnormal termination.

97
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-14 b) Test of interface verification

During the course of system development, tests are conducted in the sequence of unit test →
integration test → system test → operational test. In b) integration test, programs are combined and
verification of whether interfaces between modules and between subsystems, such as procedure and
format of transferring data, are functioning correctly is performed.
a) An operational test is a test where a user department takes the center stage and a test is
conducted using the same system and data as the production environment. An operational test
checks whether the new system can be effectively used for business operations.
c) A system test is a test that verifies from several angles whether all programs for which an
integration test is completed functions correctly as per the design document. For the entire
system, function test, load test, performance test, and operability test are conducted.
d) A unit test is conducted for individual modules. A unit test verifies whether a particular module
is functioning properly and whether the logical structure of the program is correct.

Q4-15 c) Substitute test module for a lower level module

In top-down testing, when a lower level module is not yet integrated (because it is not complete or
not in the scope of test), a test module that is used as a substitute for such a lower level module is
called a stub. Therefore, c) is the correct answer.
a) An emulator is a mechanism that allows a computer to execute the program of another computer
that has a different instruction format. It is normally implemented as software.
b) Simulator refers to software or hardware that simulates all or a part of a computer system.
d) In a bottom-up system, a driver is a test module used as a substitute of an upper level module not
yet integrated.

98
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-16 a) Tests conducted in system test process

In software development based on the waterfall model, testing is conducted in the sequence of unit
(module) test, integration test, system test, and operational test. a) Load test is conducted in the
system test phase. In a load test, a load that is equivalent to or higher than the data processing
volume at the peak time of operation is applied in order to verify that a system operates normally.
b) Conducted in the integration test phase.
c) Conducted in the unit test phase.
d) A regression test is a test for checking whether modification of a program adversely affects any
other parts than the corresponding function, and it is conducted in the maintenance phase.

Q4-17 a) Test data to be prepared for system qualification test

A system qualification test, sometimes simply called as a system test, is conducted after a system
developer has conducted a unit test and an integration test. Therefore, when it is required, users also
participate in this test to check that the entire system is working as per the specifications. A system
qualification test includes a function test to check whether all the required functions are included or
not, a performance test to check the processing capacity (throughput) and response time, an
exception processing test for error and data failure, and a load test to check that the system can
handle large volume of data processing. The data used in actual operations and the data handled by
exception processing are also required, so a) is the correct answer. In addition, a system qualification
test also includes an operability test to check usability for users. Also, the term “system qualification
test” is one of the processes of Common Frame 2007 (SLCP-JCF2007), and was previously called a
“system test.”
b) A test concerning the interface between software units is called an integration test, and is
conducted before a system test. A system is tested for errors using data not defined in the
interfaces between software units.
c) and d) Data that covers all branches and executes all instructions within a software unit one or
more times is used in a unit test conducted as per the white box test method. A unit test
verifies whether each software unit satisfies the relevant functions in the program
specifications, and it checks if there are any logical errors inside a software unit.
Moreover, in the terminology of testing techniques, c) is the test data that is used in a
decision condition (or branch) coverage, while d) is the test data that is used in statement
coverage.

99
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-18 a) Objective of system migration test

A migration test is a test to check that after a new system is developed, changeover (migration)
from the old system to the new system can be performed smoothly. A migration test is not a test of
the new system itself. Since it is also conducted to check the process for system changeover in
advance from the viewpoint of safety and efficiency, a) is the correct answer.
b), c), and d) are correct in terms of description concerning the test and performance verification
conducted prior to migration test. However, these are not the main objectives of migration test.

Q4-19 a) Test process quality control chart

From the quality control chart of the testing phase that is provided here, it is obvious that the
number of test items not executed is more than expected, and that the number of errors detected is
also more than expected.
Since the increase in the number of errors detected is large in comparison with the decline in the
number of test items not executed, there are quality problems, and it is highly possible that several
elementary errors are present in the stage prior to starting the current test. Therefore, in addition to
investigating the areas that have many defects and taking the necessary measures, the previous
phases should be reviewed as required. Accordingly, a) is the appropriate description.
b) As the number of test items not executed is higher than the expected number, deficiencies in the
test environment and a lack of development personnel may be the issues. However, since the
number of errors detected is higher than the expected number, it is highly possible that there are
many elementary errors. It is incorrect to assume that there are no quality problems.
c) The number of test items not executed is higher than the expected number, and therefore it is
incorrect to say that execution of test items is fast.
d) There can be two ways of looking at a situation where error detection is progressing better than
expected. First, the quality of test items is high, and with fewer test items, more errors are
detected than the expectation, in other words, testing is progressing efficiently. Second, the
quality of software is poorer than expected, and therefore many unexpected elementary errors
have occurred. Even if it is the former, it should be verified that this is actually the case, rather
than just thinking optimistically as attention is not required immediately. Moreover, as a result of
verification, if it is found that testing is progressing efficiently, “keeping track of the progress of
error resolution is necessary so that unsolved bugs do not remain unaddressed for a long time” is
appropriate.

100
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-20 d) Relationship between scale and workload in software development

For the relationship between the development scale and workload (i.e., person-hours) in software
development, it is known that an increase in development scale will generally increase the
person-hours of development exponentially; that is, as the development scale increases, the increase
in person-hours of development is dramatically huge. Therefore, the graph shown in d) is correct.
In addition, a) is incorrect because an increase in person-hours of development has slowed down
at the middle point, while b) and c) are incorrect because the increase in person-hours of
development is slowing down as the development scale increases.

Q4-21 a) CMMI

CMMI (Capability Maturity Model Integration) is essentially a model developed by adding an

interface with hardware development to CMM (process maturity model), which is for software
development. None of the answers describes the details entirely. However, description of a) is an
explanation of CMM, which forms the basis of CMMI. As none of the answers from b) through d)
are related to CMMI, the correct answer is a). In addition, CMM is a process evaluation method
defined by Software Engineering Institute of Carnegie Mellon University of the United States. Like
CMM, in CMMI also, maturity of development process is evaluated with the following 5-stage
evaluation criteria.
・Level 1: The initial stage where no processes are established.
・Level 2: Although individual processes are managed, they are not organized.
・Level 3: The experience of each individual is collected and documented, and a consistent
standard processes are established for the organization.
・Level 4: Standardized processes are quantitatively measured and analyzed.
・Level 5: Standardized processes are optimized and improved according to the difference in
technology and requirements of the environment.
b) “Process model for software development” refers to a model like the waterfall model and growth
model, where the method for proceeding with development work is modeled.
c) “Common frame for software-based system development and transactions” stipulates the
mechanism of a common software lifecycle process for clear specification development and
transactions. At present, the revised SLCP-JCF 2007 is the latest version.
d) No such model exists.

101
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-22 a) Buyer’s responsibilities in software transactions of JIS X 0160

JIS X 0160 was formed by the JIS standardization of international standards (ISO/IEC 12207)
concerning software lifecycle process. SLCP-JCF98 (common frame) was formed by adding
definitions of work items to JIS X 0160 after consideration of local circumstances in Japan.
As in the description in the question, both JIS X 0160 and SLCP-JCF98 stipulate supplier’s
responsibilities as well as a buyer’s responsibilities. Some of the main responsibilities of a buyer are
“defining and documenting requirements,” “definition of acceptance criteria and conditions,”
“monitoring suppliers, such as joint review and cooperation with the supplier” and so on. Therefore,
a) is the correct answer.
Answers b) through d) are not defined as buyer’s responsibilities.
b) Depending on the contract, a supplier’s responsibility may include operations up to system
operations.
c) In simple terms, “product defect” refers to a bug. Even if the responsibility of correcting this
defect lies with the supplier, it is clear that it will not be the buyer’s responsibility in most cases.
d) An internal quality audit is conducted on the supplier side, and its responsibility lies with the
supplier, not the buyer.

Q4-23 c) Extracting program specifications and desing from code

The technique of extracting information about the design and specifications of a program by
analysis of the source code or the object code is called reverse engineering. Therefore, c) is the
correct answer.
a) Reengineering refers to revision of the existing organization structure and rules, and redesign of
the business flow and job duties, management mechanism, or information systems for corporate
restructuring from the viewpoint of processes.
b) In Japan, restructuring is generally considered to be a reduction of the workforce. In correct
terms, it not only means a reduction of the workforce, but also includes transforming the
management structure by rearranging the business structure.
d) Refactoring refers to a manual modification of source code so that it does not affect or change
the execution results of the program.

102
 Morning Exam Section 4 Development Technology Answers and Explanations

Q4-24 a) Specifications of componentization of a Java program

“JavaBeans” refers to the specifications of software components that can be used from a Java
program. Individual software components are called Beans. Just as coffee is made from many beans,
this originated from the concept of creating a Java program using components (JavaBeans) that can
be reused. Therefore, a) is the correct answer.
b) JavaScript is a scripting language developed based on Java. It is a different language to Java.
c) Java application refers to a program developed in Java.
d) Java applet refers to a small program written in Java. It is downloaded from the server and then
executed on the client side. Applet is a composite word of application and –let, and it means a
“small application.”

Q4-25 c) Maintenance of development environment of embedded system

The most important point to be considered concerning maintenance of a development environment

after product development is the consideration of reuse during the product maintenance phase, and
maintenance of the development environment as it is in the development phase so that it can be used
without any delay when required. Therefore, conducting periodic operation checks is important.
Accordingly, c) is the correct answer.
a) Updating the development environment after product development to the latest condition may
lead to a situation where the environment for generating the same modules as during
development cannot be maintained.
b) This is not appropriate in consideration of the product maintenance phase and the possibility of
the subsequent development of similar products.
d) It is difficult to assume this situation because it depends on contract terms and conditions with
the rental company, including duration.

103