CS226 COMPUTER

ORGANIZATION AND
ASSEMBLY LANGUAGE

Civilization advances by extending the

number of important operations which
we can perform without thinking about
them. Alfred North Whitehead

Instructor: Asra Masood

Lecturer
Department of Creative Technologies
FCAI
Text books:

 “Assembly Language for Intel-based Computers

”, 7th Edition, Irvine, Prentice Hall 2003

 Computer Organization and Design The

Hardware/Software Interface, 5th Edition (2014)
by David A. Patterson and John L. Hennessy
Morgan Kaufmann Publishers

Assessment:
 Assignments: 15%
 Quizzes: 15%
 Midterm test: 25%
 Final exam: 45%
COMPUTER ORGANIZATION 5th
AND DESIGN Editio
n
The Hardware/Software Interface

Chapter 1
Computer Abstractions
and Technology
 MOORE’S LAW
 Moore’s Law states that integrated
circuit resources double every two
years (18–24) months.
 Moore’s Law resulted from a 1965
prediction of such growth in IC
capacity made by Gordon Moore,
one of the founders of Intel.
 Moore’s Law graph to represent
designing for rapid change
 MOORE’S LAW
Year of introduction Transistors
4004 1971 2,250
8008 1972 2,500
8080 1974 5,000
8086 1978 29,000
286 1982 120,000
386™ 1985 275,000
486™ DX 1989 1,180,000
Pentium® 1993 3,100,000
Pentium II 1997 7,500,000
Pentium III 1999 24,000,000
Pentium 4 2000 42,000,000
 IS MOORE’S LAW ENDING?
 Intel’s former chief architect Bob Colwell
says: Moore’s law will be dead within a
decade (August 2013).
 Theoretical physicist Michio Kaku
believes Moore's Law has about
10 years of life left
before ever-shrinking transistor sizes
smack up against limitations imposed
by the laws of thermodynamics and
quantum physics (April 2013).
https://www.intel.com/content/www/us/
en/newsroom/resources/moores-law.ht
ml#gs.kawmkf
IS MOORE’S LAW ENDING? (2)
 BELL’S LAW
Bell's Law for the birth and death of computer
classes:
 Bell's law of computer classes formulated by

Gordon Bell in 1972 describes how computing

systems (computer classes) form, evolve and
may eventually die out.
 Roughly every decade a new, lower priced

computer class forms based on a new

programming platform, network, and
interface resulting in new industry.
 In 1951, men could walk inside a computer

and now, computers are beginning to “walk”

inside of us .
 BELL’S LAW

Source: B Bell, “Bell’s Law for the Birth and Death of Computer Classes”, Comms of
ACM, 2008
 GENERATION
THE 1ST
COMPUTER

Source:
COMPUTERS TODAY
 Introduction
§1.1
THE COMPUTER REVOLUTION
 Progress in computer
technology
 Underpinned by Moore’s Law
 Makes novel applications
feasible
 Computers in automobiles
 Cell phones
 Human genome project
 World Wide Web
 Search Engines

 Introduction
§1.1
THE COMPUTER REVOLUTION
(2)
 Computers in automobiles
 reduce pollution, improve fuel efficiency
via engine controls, and increase safety
through blind spot warnings, lane
departure warnings, moving object
detection, and air bag inflation to
protect occupants in a crash
 Cell phones
 more than half of the planet having
mobile phones, allowing person-to-
person communication to almost
anyone anywhere in the world
 THE COMPUTER REVOLUTION

Introduction
§1.1
(3)
 Human genome project
 a global effort to identify the estimated
30,000 genes in human DNA to figure
out the sequences of the chemical
bases that make up human DNA to
address ethical, legal, and social issues
 The cost of computer equipment to
map and analyze human DNA
sequences was hundreds of millions of
dollars. Since, costs continue to drop,
we will soon be able to acquire our
own genome, allowing medical care to
be tailored to us
 Introduction
§1.1
THE COMPUTER REVOLUTION
(4)
 World Wide Web
 has transformed our society. Web
has replaced libraries and
newspapers
 Search Engines
 As the content of the web grew in size
and in value, many people rely on
search engines for such a large part of
their lives that it would be a hardship to
go without them
 CLASSES OF COMPUTERS

 Personal
computers
 Server computers

 Supercomputers

 Embedded

computers
 CLASSES OF COMPUTERS
 Personal computers
 Computers designed for use by an
individual
 General purpose, variety of software
 Subject to cost/performance tradeoff
 Server computers
 Computers used for running larger
programs for multiple users, often
simultaneously
 Network based
 High capacity, performance, reliability
 Range from small servers to building
 CLASSES OF COMPUTERS
 Supercomputers
 Consist thousands of processors and
many terabytes of memory
 High-level scientific and engineering
calculations
 Highest capability and cost but represent a
small fraction of the overall computer
market
 Embedded computers
 Hidden as components of systems
 Largest class of computers and span the
widest range of applications

 THE POSTPC ERA
 Smart phones
represent the recent
growth in cell phone
industry, and they
passed PCs in 2011.
 Tablets are the
fastest growing
category, nearly
doubling between
2011
and 2012.
 Recent PCs and
traditional cell
phone categories
are relatively flat or
 THE POSTPC ERA
 Personal Mobile Devices (PMDs)
 Small wireless devices
 Battery operated
 Connects to the Internet
 Hundreds of dollars
 Smart phones, tablets, electronic
glasses
 THE POSTPC ERA
 Cloud computing
 Term cloud essentially used for the
Internet
 Portion of software run on a PMD
and a portion run in the Cloud
 Warehouse Scale Computers
(WSC)

Big datacenters containing 100,000
servers

Amazon and Google cloud vendors
 Software as a Service (SaaS)

Delivers software and data as a service
over the Internet
 WHAT YOU WILL LEARN
 How programs are translated
into machine language
 and how hardware executes
them
The hardware/software


interface
 What determines program
performance
 and how it can be improved
 How hardware designers
improve performance
UNDERSTANDING PERFORMANCE
 Algorithm
 determines number of operations

executed
 Programming language, compiler,
architecture
 determine number of machine

instructions executed per operation

 Processor and memory system
 determine how fast instructions are

executed
 I/O system (including OS)
 Architecture
§1.2 Eight Great Ideas in Computer
EIGHT GREAT DESIGN IDEAS
 Design for Moore’s Law

Computer designs can take years, resources
available per chip can easily double or
quadruple between start and finish of project

Anticipate where technology will be when
design finishes
 Use Abstraction to simplify design
 Hide lower-level details to offer a simpler model
at higher levels
 Make the Common Case Fast

To enhance performance better than
optimizing the rare case
 Architecture
§1.2 Eight Great Ideas in Computer
EIGHT GREAT DESIGN IDEAS
 Performance via Parallelism

A form of computation in which many
calculations are carried out simultaneously
 Performance via Pipelining

A particular pattern of parallelism

A set of data processing elements connected
in series, so that the output of one element is
the input of the next one
 Performance via Prediction

It can be faster on average to guess
and start working rather than wait
 Architecture
§1.2 Eight Great Ideas in Computer
EIGHT GREAT DESIGN IDEAS
 Hierarchy of memories

The closer to the top, the faster and more
expensive per bit of memory

The wider the base of the layer, the
bigger the memory
 Dependability via redundancy

Design systems dependable by
including redundant components

to take over when failure occurs,
and

to help detect failures
 Program
§1.3 Below Your
BELOW YOUR PROGRAM
 Application software

Written in high-level language
 System software

Compiler: translates HLL
code to machine code

Operating System:

Handling input/output

Managing memory and storage

Scheduling tasks & sharing
resources
 Hardware

Processor, memory, I/O
controllers
 LEVELS OF PROGRAM CODE
 High-level language

Level of abstraction
closer to problem
domain

Provides for
productivity and
portability
 Assembly language

Textual
representation of
instructions
 Hardware
representation

Binary digits (bits)
 Covers
§1.4 Under the
COMPONENTS OF A COMPUTER
The BIG Picture
 Same components
for all kinds of
computer

Desktop,
Server,
Embedded
 Input/Output
includes

User-interface devices

Display, keyboard,
mouse

Storage devices

Hard disk, CD/DVD,
flash
 ABSTRACTIONS
The BIG Picture
 Abstraction helps us deal with
complexity
 Hides lower-level details
 Instruction Set Architecture
(ISA) or Computer Architecture
 The hardware/software interface
 Includes instructions, registers,
memory access, I/O, and so on
 Operating system hides details of
doing I/O, allocating memory from
 A SAFE PLACE FOR DATA
 Volatile main memory
 Loses instructions and data when
power off
 Non-volatile secondary memory
 Flash memory
 Optical disk (CDROM, DVD)
 Magnetic disk
 NETWORKS
Backbone of computer systems
 Communication


Information is exchanged between computers
at high speeds
 Resource sharing

Rather than each computer having its own
I/O devices, computers on the network can
share I/O devices
 Nonlocal access

By connecting computers over long
distances, users need not be near the
computer they are using
 NETWORKS
 Local area network (LAN): Ethernet
 A network designed to carry data
within a geographically confined area,
typically within a single building
 Wide area network (WAN): the
Internet
 A network extended over
hundreds of kilometers that can
span a continent
 Wireless network: WiFi,
Bluetooth
 Performance
§1.6
DEFINING PERFORMANCE
 Which airplane has the best
performance?
Boeing 777 Boeing 777

Boeing 747 Boeing 747

BAC/Sud BAC/Sud
Concord Concord
e e
Dougla Douglas
s DC-8- DC- 8-
50 50

0 Passenger 0 Cruising Range

Capacity (miles)
1 2
0 0
0 0
Boeing Boeing 777 0
777 2
0 Boeing 747
Boeing 0 4
747 BAC/Sud 0
3 Concord 0
BAC/Sud 0 0
e
Concord 0
Douglas
e
4 DC- 8- 6
Dougla 0 500 1000 1500 50 0 100000 200000 300000
s DC-8- 0 0 400000
50 Cruising Speed 0 Passengers x
5 (mph) mph
0
0 8
0
0
0
 RESPONSE TIME AND
THROUGHPUT
 Response time

How long it takes to do a task
 Throughput

Total work done per unit time
e.g., tasks/transactions/… per hour
 How are response time and throughput
affected by

Replacing the processor with a faster version?

Adding more processors?
 We’ll focus on response time for now…
 RELATIVE PERFORMANCE
 Define Performance = 1/Execution
Time
 “XPerformance
is n timesXfaster thanY Y”
Performance
 Execution time Y Execution
time X  n
 Example: time taken to run a
program
 10s on A, 15s on B
 Execution TimeB / Execution TimeA
= 15s / 10s = 1.5 = 1½
 So A is 1½ times faster than B
 MEASURING EXECUTION TIME
 Elapsed time

Total response time, including all aspects

Processing, I/O, OS overhead, idle time

Determines system performance
 CPU time

Time spent processing a given job

Minus I/O time, other jobs’ shares

Includes user CPU time and system CPU
time

Different programs are affected differently
by CPU and system performance

Running on servers – I/O performance –
hardware and software

Total elapsed time is of interest

Define performance metric and then proceed
 CPU CLOCKING
 Operation of digital hardware
governed by a constant-rate clock
Clock
period

Clock (cycles)

Data transfer
and
computation
Update state
 Clock period: duration of a clock
cycle

e.g., 250ps = 0.25ns = 250×10–12s
 Clock frequency (rate): cycles per
second
 CPU TIME
CPU Time  CPU Clock Cycles Clock Cycle
Time

CPU Clock

 Performance can be improved by
Cycles Clock
 Reducing number
Rate of clock cycles
 Increasing clock rate
 Hardware designer must often trade off
clock rate against cycle count
 CPU TIME EXAMPLE
 Computer A: 2GHz clock, 10s CPU time
 Designing Computer B
 Aim for 6s CPU time

 Can do faster clock, but causes 1.2 × clock cycles

 How fast must Computer B clock be?

Clock B 
Clock CyclesB 1.2 Clock
CPU Time 
Rate Cycles
B
6s A
Clock CyclesA  CPU Time A Clock
RateA

 10s 2GHz 9 20 109

1.2  20 10 24
Clock RateB  
1096s
 4GHz
6s
 INSTRUCTION COUNT AND CPI
Clock Cycles  Instruction Count  Cycles Per
Instruction
CPU Time  Instruction Count  CPI Clock
Cycle Time
 Instruction
Instruction Count Count CPI
for aprogram


Determined by program,
Clock Rate ISA, and
compiler
 Average cycles per instruction

Determined by CPU hardware

If different instructions have different
CPI

Average CPI gets affected by
 CPI EXAMPLE
 Computer A: Cycle Time = 250ps, CPI
= 2.0
 Computer B: Cycle Time = 500ps, CPI

= 1.2
 Same ISA

Which
 CPU TimeisA faster? by how
 Instruction Countmuch?
 CPIA  Cycle
A is faster…
Time A

 I 2.0  250ps  I 500ps

CPU TimeB  Instruction Count  CPIB 

CPU TimeB I 600ps … b y this
Cycle TimeB   much

 I1.2  500ps  I 600ps

CPU Time
 CPI IN MORE DETAIL
 If different instruction classes take
different numbers of cycles
n

Clock Cycles   (CPIi Instructio n

Counti )
 Weighted average
i1

CPI Clock Cycles n

 Instruction
CPI   CPI i 
 Instruction i1
 Instruction
Count i 
Count  Count 
Relative
frequency
 CPI EXAMPLE
 Alternative compiled code sequences
using instructions in classes A, B, C
Class A B C
CPI for class 1 2 3
IC in sequence 1 2 1 2
IC in sequence 2 4 1 1

 Sequence 1: IC =  Sequence 2: IC
5 = 6
Clock Cycles

Clock Cycles = 4×1 + 1×2 +
= 2×1 + 1×2 + 1×3
2×3 =9
= 10 
Avg. CPI = 9/6 =
 1.5
 PERFORMANCE SUMMARY
The BIG Picture

Instructions Clock Cycles

CPU Time   
Seconds Program Instruction
Clock Cycle

 Performance depends on
 Algorithm: affects IC, possibly CPI
 Programming language: affects IC, CPI
 Compiler: affects IC, CPI
 Instruction set architecture: affects IC,
 Wall
§1.7 The Power
POWER TRENDS

Clock rate and Power for Intel x86 microprocessors over eight
generations
 In CMOS IC technology
Power  Capacitive load Voltage 2

Frequency
×30 5V → 1V ×1000
 REDUCING POWER
 Suppose a new CPU has
 85% of capacitive load of old CPU
 15% voltage and 15% frequency
reduction
Pnew Cold  0.85  (Vold2 0.85)2 Fold  0.854 
Pold  C  V F
old old 0.52
old
0.85

 The power wall
 We can’t reduce voltage further
 We can’t remove more heat
 How else can we improve
performance?
 Multiprocessors
§1.8 The Sea Change: The Switch to
UNIPROCESSOR PERFORMANCE
Technology
driven

Advanced architectural and organizational ideas

Constrained by power, instruction-level parallelism,

memory latency
 MULTIPROCESSORS
 Multicore microprocessors
 More than one processor per chip
 Requires explicitly parallel
programming
 Compare with instruction level
parallelism

Hardware executes multiple instructions
at once

Hidden from the programmer
 Hard to do (Why?)

Programming for performance

Load balancing
 Remarks
§1.9 Concluding
CONCLUDING REMARKS
 Cost/performance is improving
 Due to underlying technology
development
 Hierarchical layers of abstraction
 In both hardware and software
 Instruction set architecture
 The hardware/software interface
 Execution time: the best
performance measure
 Power is a limiting factor
 Use parallelism to improve
performance
 ACKNOWLEDGEMENT
•The slides are adopted from Computer Organization
and Design, 5th Edition by David A. Patterson and
John L. Hennessy 2014, published by MK (Elsevier)