Chapter 5

Computer Systems Organization

Introduction

 This chapter changes the level of abstraction and

focuses on a higher level of computer system construction
 Focus on functional units and computer organization
 A hierarchy of abstractions hides unnecessary details
 Change focus from transistors to gates and to circuits as
the basic unit
 Discusses in detail the Von Neumann architecture
What do
you
drive?
What type of
computer?
 Find three different computers online (laptops or
Find desktops).

Lets Take Record the following details for each:

Time to
• Brand & Model
Record • Operating System (Windows, macOS, Linux, etc.)
• RAM (8GB, 16GB, etc.)
• Storage Type & Size (HDD vs. SSD, 256GB, 1TB, etc.)

Look
• Price (optional but insightful)

Include Include your own

 Get in small Groups

Which
Each student presents
computer their list.

is best?
Groups vote on the
best overall
computer based on
specs and price.
 Which computer would
be best for a student? A
gamer? A business
user?

How does storage type

Which (HDD vs. SSD) impact
performance?
computer
is best? Why does RAM matter?

How does their personal

computer compare?
Components of
a Computer
System
Von Neumann
architecture is the
foundation for nearly all
modern computers
• The four major subsystems
of the Von Neumann
architecture
• Memory
• Input/output
• Arithmetic/logic unit (ALU)
• Control Unit
• ALU and control unit are
often bundled inside the
central processing
unit (CPU)
Computer
Crash?
Memory

 Random Access Memory  Read Only Memory

 Most important  Only read data
 Causes error if not  Storage medium
enough
  When turn off it will
Stores data temporary
remain
 Volatile
 Bootstrap (to start
 All the data and
information stored will the computer)
be deleted  ROM is lower and
 Sram – interlocking much cheaper
transistors (does not
 Does not need
need refreshing)
power to store data
 Storage capacity is 1 to
256 GB per chip
The Components of a Computer
System Memory and Cache
 Memory: functional unit where data and instructions are
stored/retrieved
 Random access memory (RAM)
 Organized into cells, each given a unique address
 Equal time to access any cell
 Cell values may be read and changed
 The time it takes to fetch/store contents of a cell is the same
for all cells in memory
 Read-only memory (ROM): A type of RAM with
prerecorded information that cannot be modified or
changed
 Cell size/memory width is typically 8 bits (byte)
 Maximum memory size/address space is 2N, where N is
length of address
 The
Components
of a
N Maximum Memory Size Computer
(2N) System
16 65,536 Memory and
20 1,048,576 Cache
22 4,194,304
24 16,777,216
32 4,294,967,296
40 1,099,511,627,776
50 1,125,899,906,842,624
 Fetch: retrieve from memory
(nondestructive fetch)
 Store: write to memory (destructive
store)
The
 Memory access time
Components
 Time required to fetch/store
of a
 Modern RAM requires 5-10 Computer
nanoseconds System
 Memory address register (MAR) holds
memory address to access
Memory and
 Memory data register (MDR) receives
Cache
data from fetch and holds data to be
stored
The Components of a Computer
System Memory and Cache
 Memory system circuits: decoder and fetch/store
controller
 Decoder converts MAR into signal to a specific memory
cell
 One-dimensional versus two-dimensional memory
organization
 Fetch/Store controller ► traffic cop for MDR
 Takes in a signal that indicates fetch or store
 Routes data flow to/from memory cells and MDR
The Components of a Computer
System Memory and Cache
 RAM speeds increased more slowly than CPU speeds
 Cache memory is fast but expensive
 Built into the CPU for fast access times
 5 to 10 times faster
 Principle of locality
 Values close to recently accessed memory are more likely to
be accessed
 Load neighbors into cache and keep recent values there
 Cache hit rate: percentage of times values are found in
cache
The Components of a Computer System
Input/Output and Mass Storage

Input/output

• Input/output (I/O) connects the processor to the outside

world
• Humans: keyboard, monitor, etc.
• Data storage: hard drive, DVD, flash drive
• Other computers: network
RAM

• RAM = volatile memory (gone without power)

Mass

• Mass storage systems = nonvolatile memory

• Direct access storage devices (DASDs)
• Sequential access storage devices (SASDs)
DASDs (Direct Access Storage
Device)

Disks: Hard drives/optical

media (CDs/DVDs)

• Sectors: fixed size segments of tracks,

unit of retrieval
• Time to retrieve data based on
• Seek time
• Latency
• Transfer time
flash memory and solid-state
drives (random access mass
storage)
SSD?
The Components of a Computer
System I/O and Mass Storage
DASDs and SASDs are orders of
magnitude slower than RAM:
(microseconds or milliseconds).

I/O Controller manages data

transfer with slow I/O devices,
freeing processor to do other
work.

Controller sends an interrupt

signal to processor when I/O
task is done.
The Components of a Computer
System The Arithmetic/Logic Unit
 ALU is part of the processor
 Contains circuits for arithmetic
 Addition, subtraction, multiplication, and division
 Contains circuits for comparison and logic
 Equality, and, or, not
 Contains registers: high-speed, dedicated memory
connected to circuits
 Data path: how information flows in the ALU
 From registers to circuits
 From circuits back to registers
The Components of a Computer
System The Control Unit

Stored program characteristic

Programs are encoded in binary and stored in computer’s memory

Control unit fetches instructions from memory,

decodes them, and executes them

Instructions encoded
Operation code (op code) tells Addresses tell which memory
which operation addresses/registers to operate on
The Components of a Computer
System The Control Unit

Machine Instruction set

language
Binary strings encode Instructions implemented by a
instructions particular chip
Instructions can be carried out Each kind of processor has a
by hardware different instruction set
Sequences of instructions (speaks a different language)
encode algorithms
 Instruction set examples:
 Data transfer, e.g., move data from
The memory to register

Components  Arithmetic, e.g., add, but also AND

 Comparison: compare two values
of a  Branch: change to a nonsequential
Computer instruction
System The  Branching allows for conditional and loop
Control Unit 
forms
E.g., JUMPLT a = If previous comparison
of A and B found A < B, then jump to
instruction at address a
The Components of a Computer
System The Control Unit

Control unit contains:

 Program counter (PC) register: holds address of next
instruction
 Instruction register (IR): holds encoding of current
instruction
 Instruction decoder circuit
 Decodes op code of instruction and signals helper circuits,
one per instruction
 Helpers send addresses to proper circuits
 Helpers signal ALU, I/O controller, and memory
Putting the Pieces Together—the
Von Neumann Architecture
Fetch/
Combine Fetch Decode Execute
decode

Combine Fetch/ Fetch Decode Execute

previous decode/ phase: get phase: phase:
pieces: execute next instruction different
instruction decoder for each
Von phase into gets op instruction
Neuman • Machine
memory code
n repeats until
HALT
machine instruction
or error
• Also called
Von
Neumann
cycle
  Problems to solve are always larger
 Computer chip speeds no longer increase
exponentially
 Reducing size puts gates closer together,
Non–Von faster
 Speed of light pertains to signals through
Neumann wire
Architectures  Cannot put gates much closer together
 Heat production increases too fast
 Von Neumann bottleneck: inability of
sequential machines to handle larger
problems
  Non–Von Neumann architectures
 Other ways to organize computers
 Most are experimental/theoretical,
EXCEPT parallel processing
Non–Von  Parallel processing
Neumann  Many processing units operating at the
same time
Architectures
 Supercomputers (in the past)
 Desktop multicore machines and “the
cloud” (in the present)
 Quantum computing (in the future)
Non–Von Neumann
Architectures
MIMD parallel processing
 Multiple instruction stream/Multiple data streams
 Cluster computing
 Multiple, independent processors
 Each ALU operates on its own data
 Scalable (can increase or decreate in performance due to
changes)
 Each processor can operate independently
 On its own data
 On its own program
 At its own rate
 Varieties of MIMD systems
 Special-purpose systems: newer
supercomputers
 Cluster computing: standard
machines communicating over LAN
or WAN
Non–Von  Grid computing: machines of
varying power, over large
Neumann distances/Internet

Architectures
 Examples
 SETI project
 BOINC at Berkley
 Hot research area ► parallel
algorithms
 Need to take advantage of all this
processing power
 We must abstract in order to manage
system complexity—no more writing
instructions in machine language.

Von Neumann architecture is standard for

modern computing.

Summary Von Neumann machines have memory, I/O,

ALU, and control unit; programs are stored
in memory; execution is sequential unless
program says otherwise.

Memory is organized into addressable

cells; data is fetched and stored based on
MAR and MDR; uses decoder and
fetch/store controller.