You are on page 1of 19

Princess Sumaya Univ.

Computer Engineering Dept.

‫ بســام كحـالــه‬.‫د‬
Dr. Bassam Kahhaleh
Princess Sumaya Univ.
Computer Engineering Dept.

Chapter 1:
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Technology
Year Technology Used in Computers Relative Perf/Unit Cost
1951 Vacuum tube 1
1965 Transistor 35
1975 Integrated circuit 900
1995 Very Large-Scale Integrated Circuit 2,400,000
2005 Ultra Large-Scale Integrated Circuit 6,200,000,000

3 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Technology
1985 2006
A computer with certain performance,
$1,000,000 $500
main memory, disk storage

4 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Classes of Computers
1960s 1970s 1980s 1990s

Large Minicomputer Desktop Servers


&
Mainframes Supercomputers Computers (Internet
&
WWW)
 Desktop
● Features, Price & Performance
Amazon.com (fall quarter of 2005) had
 Servers an average revenue per hour of $1.35
million
● Reliability
Computing capacity, Memory,
● Scalability Storage, & I/O Bandwidth

● Efficient Throughput (Performance: TPM)

5 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Classes of Computers

 Desktop
 Servers
 Embedded Computers
● Price
● Real-time Performance
● Minimized Memory
● Minimized Power

6 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Computer Performance

 Response Time
The time it takes to do a task  Execution Time

 Throughput
The total amount of work done in a given time

 Difference?
Task A Task B

Calculate Calculate Calculate Calculate


Read File Save File Read File Save File

Time
0t 1 2 3 4 5 6 7 8 9 10

7 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Computer Performance

 Response Time
The time it takes to do a task  Execution Time

 Throughput
The total amount of work done in a given time

 Difference?
Task A
Task B
Calculate Calculate Calculate Calculate
Read File Read File Save File Save File

Time
0t 1 2 3 4 5 6 7 8 9 10

8 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

CPU Performance

 Performance = 1 / Execution Time


 Relative Performance:
Performance Computer X Execution time Y
●n = ──────────────── = ───────────
Performance Computer Y Execution time X

 CPU Performance
●CPU Execution time
= CPU Clock Cycles × Clock Cycle time
CPU Clock Cycles
= ─────────────
Clock Rate

9 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

CPU Performance
Exercise:
A program takes 10 seconds to run on a 2 GHz CPU. The same
program on another CPU would take 20% extra clock cycles,
yet it finishes in 6 seconds. What is the other CPU clock rate?

10 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Instruction Performance

 Clocks Per Instruction, CPI


The average number of clock cycles each instruction
takes to execute.
Exercise: Computer Clock Cycle Time CPI
Which computer is faster?A 250 picoseconds 2
B 500 picoseconds 1.2

# of instructions in a program =
CPU-A: CPU Execution Time = × × ps = ps
CPU-B: CPU Execution Time = × × ps = ps
Computer A is ( / )= times faster than B

11 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Instruction Performance
Exercise:
Instruction Class
Given 3 groups of instructions: A, B and A B C
C, it takes different clock cycles to execute CPI 1 2 3
an instruction within each group.
Instruction Count
Given the shown instruction mix, which Code
A B C
code sequence is faster to execute?
Seq1 2 1 2
Seq2 4 1 1

Seq1: CPU Execution Time = × + × + × = cycles


Seq2: CPU Execution Time = × + × + × = cycles
Seq2 is / = times faster than Seq1
Seq1 average CPI = = cycles / instruction
Seq2 average CPI = = cycles / instruction
12 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

The Power Wall

 Voltage, Capacitive Load & Frequency

13 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Multi-Core

Limitation in:
Power
Parallelism
Technology Architectural & Memory
Organizational
Improvements

14 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Chip Manufacturing

15 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Die Cost

Cost per Wafer


Die Cost = ──────────────
Dies per Wafer × Yield

Wafer Area
Dies per Wafer = ───────
Die Area
1
Yield = ─────────────────────
Defects per Area × Die Area 2
(
1+ ───────────────── )
2

16 / 16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Amdahl’s Law

 When introducing an improvement,


Execution Time is divided into 2 parts:
● Affected by the improvement
● Not affected
Execution Time Execution Time Affected
Execution
After = ──────────────── + Time
Improvement Amount of Improvement
Unaffected
Add Multiply
Example: ─── = ─── +
Program Execution Time 20 seconds 80 seconds

How much improvement is required for the multiply 17 / 16


Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

 Exercise 1.3
 Exercise 1.4
 Exercise 1.5
 Exercise 1.6
 Exercise 1.11
 Exercise 1.13
 Exercise 1.14
 Exercise 1.15
 Exercise 1.16
Princess Sumaya University 22540 – Computer Arch. & Org (2) Computer Engineering Dept.

Chapter 1

You might also like