Scribd
Upload
262 views6 pages

Exercises With Solutions

This document contains solutions to 10 exercises about computer architecture from a textbook. It discusses the performance of two processors (P1 and P2) on two machines (M1 and M2). Some key points extracted: - M2 is 4/3 times as fast as M1 for P1, and P1 is 2 times as fast as M2 - M2 runs programs 4/3 as fast as M1 but costs 8/5 as much, so M1 has better value - M2 can run 120 iterations of program P2 within 3600 seconds while M1 can only run 80 iterations - Program P2 has a peak performance of 3000 MIPS while P1 is 4000 MIPS

Uploaded by

Serxhino Korreshi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
262 views6 pages

Exercises With Solutions

This document contains solutions to 10 exercises about computer architecture from a textbook. It discusses the performance of two processors (P1 and P2) on two machines (M1 and M2). Some key points extracted: - M2 is 4/3 times as fast as M1 for P1, and P1 is 2 times as fast as M2 - M2 runs programs 4/3 as fast as M1 but costs 8/5 as much, so M1 has better value - M2 can run 120 iterations of program P2 within 3600 seconds while M1 can only run 80 iterations - Program P2 has a peak performance of 3000 MIPS while P1 is 4000 MIPS

Uploaded by

Serxhino Korreshi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Islamic University Gaza

Engineering Faculty
Department of Computer Engineering
ECOM 3010: Computer Architecture Discussion

Chapter 4
Exercises with solutions

Eng. Eman R. Habib

November, 2013
2 Computer Architecture Discussion

Exercise 1: (4.1 from book)

For P1:
Performance M2/Performance M1 = Execution time M1/ Execution time M2
= (2 sec/1.5 sec) = 4/3
M2 is 4/3 times as fast asM1.

For P2:
Performance M1/Performance M2 = Execution time M2/ Execution time M1
= (10 sec/5 sec) = 2
M1 is 2 times as fast asM2.

Exercise 2: (4.2 from book)

For P1 on M1:
The execution rate = number of instructions / time
= 5 109 instructions /2 seconds = 2.5 109 IPS

For P1 on M2:
The execution rate = number of instructions / time
= 6 109 instructions /1.5 seconds = 4 109 IPS
3 Computer Architecture Discussion

Exercise 3: (4.3 from book)

M2 runs 4/3 as fast as M1


But M2 costs 800/500 =8/5 as much.
As 8/5 is more than 4/3.
M1 has the better value.

Exercise 4: (4.6 from book)

Running P1 1600 times on M1 = 1600 2 = 3200 seconds


Running P1 1600 times on M2 =1600 1.5 = 2400 seconds
Running P2 on M1 = 3600 - 3200 = 400 seconds
Running P2 on M2 =3600 - 2400 = 1200 seconds
M1 can run (400 seconds / (5 seconds/iteration)) = 80 iterations of P2.
M2 can run (1200 seconds / (10 seconds/iteration)) = 120 iterations of P2.
Thus M2 performs better on this workload.
Cost-effectiveness:
For M1 it costs ($500 / (80 iterations/hour)) = $6.25 per (iteration/hour)
For M2 it costs ($800 / (120 iterations/hour)) = $6.67 per (iteration/hour)
Thus M1 is most cost-effective.

Exercise 5: (4.7 from book)

a. CPU Time = clock cycle time number of instructions CPI


= 1/(5 109) (Sec/cycle) (7.5 109 instructions) (0.8 cycles/instruction) = 1.2 sec
b. P received 1.2 seconds/3 seconds or 40% of the total CPU time.
4 Computer Architecture Discussion

Exercise 6: (4.8 from book)

The ideal instruction sequence for P1 is one composed entirely of instructions from class A
(which have CPI of 1).
So P1's peak performance is (4 109 cycles/second) / (1 cycle/instruction) = 4000 MIPS.
The ideal sequence for P2 contains only instructions from A, B, and C (which all have a CPI of 2).
So P2's peak performance is (6 109 cycles/second) / (2 cycles/instruction) = 3000 MIPS.

Exercise 7: (4.9 from book)

The average CPI of P1 is (1 2 + 2 + 3 + 4 + 3) / 6 = 7/3.


The average CPI of P2 is (2 2 + 2 + 2 + 4 + 4) / 6 = 8/3.
Performance p2/Performance p1 = Execution time p1/ Execution time p2
= ((7/3 cycles/instruction) / (4 109 cycles/second))
/ ((8/3 cycles/instruction) / (6 109 cycles/second)) = 1.3125
P2 is 1.3125 times faster than P1.
5 Computer Architecture Discussion

Exercise 8: (4.14 from book)

The total execution time for the machines are as follows:


Computer A = 2 + 20 + 200 seconds = 222 seconds
Computer B = 5 + 20 + 50 seconds= 75 seconds
Computer C = 10 + 20 + 15 seconds = 45 seconds

Thus computer C is faster.


It is 75/45 = 5/3 times faster than computer B
And 222/45 = 74/15 times faster than computer A.

Exercise 9: (4.17 from book)

CPI = Execution time (second) clock rate (cycles/second) / number of instructions (Instructions)
CPI of P1 on M1 is
(2.0 seconds) (4 109 cycles/second) / (5 109 instructions) = 1.6 CPI
CPI of P1 on M2 is
(1.5 seconds) (6 109 cycles/second) / (6 109 instructions) = 1.5 CPI.
6 Computer Architecture Discussion

Exercise 10: (4.18 from book)

We have the CPI, the clock rate, and the total execution time, and we're trying to find the total
number of instructions. Using the following equation:
Instructions = Execution time (second) clock rate (cycles/second) / CPI (Cycles/instruction)
instructions in P2 on M1 is
(5.0 seconds) (4 109 cycles/second) / (1.6 Cycles/ instructions) = 12.5 109 instructions
instructions in P2 on M2 is
(10.0 seconds) (6 109 cycles/second) / (1.5 Cycles/ instructions) = 40 109 instructions

Best Wishes

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy