Docsity
Docsity
Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity

Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan

Guidelines and tips
Guidelines and tips
Sell on Docsity
Sell on Docsity
Docsity AI
Docsity AI
Log inSign up

Prepare for your exams

Study with the several resources on Docsity

Find documents

Prepare for your exams with the study notes shared by other students like you on Docsity

Search for your university

Find the specific documents for your university's exams

Docsity AINEW

Summarize your documents, ask them questions, convert them into quizzes and concept maps

Explore questions

Clear up your doubts by reading the answers to questions asked by your fellow students

Earn points to download

Earn points by helping other students or get them with a premium plan

Share documents
download point icon20 Points

For each uploaded document

Answer questions
download point icon5 Points

For each given answer (max 1 per day)

All the ways to get free points
Get points immediately

Choose a premium plan with all the points you need

Study Opportunities

Choose your next study program

Get in touch with the best universities in the world. Search through thousands of universities and official partners

Community

Ask the community

Ask the community for help and clear up your study doubts

Free resources

Our save-the-student-ebooks!

Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors

CS/COE1541 Assignment 4: Computer Architecture - Amdahl's Law, HDD Performance, and Concur, Assignments of Computer Science

University of Pittsburgh (Pitt) - Medical Center-Health SystemComputer Science
Prof. Sangyeun Cho

The university of pittsburgh, department of computer science, cs/coe1541 - intro. To computer architecture assignment 4. The assignment includes questions on amdahl's law, hard disk drive performance, and concurrent execution. Students are required to calculate the overall speedup, plot a graph, discuss the differences between shared memory and message passing programming models, analyze the impact of various hard disk drive features, and determine possible execution interleaving orders and output combinations for concurrently executed programs.

Typology: Assignments

Pre 2010

Uploaded on 09/02/2009

koofers-user-6x3-1
koofers-user-6x3-1 🇺🇸

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
1/2
UniversityofPittsburgh
DepartmentofComputerScience
CS/COE1541–Intro.toComputerArchitecture
Assignment#4(DueonWednesdayApril15)
NOTE:Thisassignmentshouldbedonebyyourself.Nolatesubmissionisaccepted.
1.[50](Amdahl’sLaw)Yourprogramcanbedecomposedintoa“serial”sectionanda“parallel”section.Theserial
section’sportionintheprogramiss(s<=1)andtheparallelsection’sportionisp(p<=1).Notethats+p=1.Assume
thattheparallelsectioncanberunbyanarbitrarynumberofprocessors,sayNprocessors,andyieldsa
commensuratespeedupofN.
(a)[30]WhatistheoverallspeeduponewouldgetfromtheprogramifN=8?
(b)[20]Plotagraphshowingtherelationshipbetweenp(sameas(1–s)),N,andspeedup.TheXaxisisthevalueof
p.Therangetoshowis0to1.TheYaxisistheoverallprogramspeedupyouwillget.Employ4lineseach
representingN=1,N=2,N=4,andN=8.
2.[30]Discussthedifferencesbetweenthesharedmemoryprogrammingmodelandthemessagepassing
programmingmodel.Chooseonemodelforyourselfandexplainwhyyoupreferthemodelovertheother.
3.[20]Wehaveaharddiskdrive.Someofthemostimportanttechnicalfeaturesofthisharddiskdriveare:
7,200RPMrotatio nspeed,200GBcapacity,andmaximum100MB/stransferratetothehostcomputer.
Thetimetogetdatafromthisharddisktothehostcomputer’smainmemoryisgivenasfollows:
TIME=controlleroverhead+seektime+rotationaldelay+readtime+transfertime
(a)[5]Describetheimpactofimprovingtherotationspeedto14,400RPMonTIME.
(b)[5]Describetheimpactofimprovingtherecordingdensity(bits/inch)onTIME.
(c)[5]Describetheimpactoffinertrackpitch(i.e.,shortertracktotrackdistance)onTIME.
(d)[5]Describetheimpactofstripingdataovermultiplesurfaces(assumingwehavemultipleplatters)withinthis
harddiskdriveonTIME.

pf2

Partial preview of the text

Download CS/COE1541 Assignment 4: Computer Architecture - Amdahl's Law, HDD Performance, and Concur and more Assignments Computer Science in PDF only on Docsity!

University of Pittsburgh

Department of Computer Science

CS/COE1541 – Intro. to Computer Architecture

Assignment #4 (Due on Wednesday April 15)

NOTE: This assignment should be done by yourself. No late submission is accepted.

  1. [50] (Amdahl’s Law) Your program can be decomposed into a “serial” section and a “parallel” section. The serial section’s portion in the program is s (s<=1) and the parallel section’s portion is p (p<=1). Note that s + p = 1. Assume that the parallel section can be run by an arbitrary number of processors, say N processors, and yields a commensurate speedup of N. (a) [30] What is the overall speedup one would get from the program if N = 8? (b) [20] Plot a graph showing the relationship between p (same as (1 – s)), N, and speedup. The X axis is the value of p. The range to show is 0 to 1. The Y axis is the overall program speedup you will get. Employ 4 lines each representing N = 1, N = 2, N = 4, and N = 8.
  2. [30] Discuss the differences between the shared memory programming model and the message passing programming model. Choose one model for yourself and explain why you prefer the model over the other.
  3. [20] We have a hard disk drive. Some of the most important technical features of this hard disk drive are: 7,200RPM rotation speed, 200GB capacity, and maximum 100MB/s transfer rate to the host computer. The time to get data from this hard disk to the host computer’s main memory is given as follows: TIME = controller overhead + seek time + rotational delay + read time + transfer time (a) [5] Describe the impact of improving the rotation speed to 14,400RPM on TIME. (b) [5] Describe the impact of improving the recording density (bits/inch) on TIME. (c) [5] Describe the impact of finer track pitch (i.e., shorter track to track distance) on TIME. (d) [5] Describe the impact of striping data over multiple surfaces (assuming we have multiple platters) within this hard disk drive on TIME.
  1. [BONUS PROBLEM, 30] Consider the concurrent execution of two programs by two processors with a shared memory. Assume that A, B, C, D are initialized to 0 and that a Print statement prints both arguments indivisibly at the same cycle. The output forms a 4 ‐tuple as either ADBC or BCAD. P0 P a. A = 1; d. C = 1; b. B = 1; e. D = 1; c. Print A, D; f. Print B, C (a) [10] List all execution interleaving orders of six statements which will preserve the individual program order. (b) [10] Assume program orders are preserved and all memory accesses are atomic; i.e., a store by one processor is immediately seen by all the remaining processors. List all the possible 4 ‐tuple output combinations. (c) [10] Assume program orders are preserved but memory accesses are non‐atomic; i.e., a store by one processor may be buffered so that some other processors may not immediately observe the update. List all possible 4 ‐tuple output combinations. Submit your work at the class or directly to the mailbox of the instructor (box #276), located in the mail room on 5th floor, SENSQ.