D R A F T

CS 691 Cluster Computing

West Virginia University

Spring semester 2002

Syllabus

 

Beowulf, the epic poem

 

<put in excerpt from poem and link to full text>

comment on significance of poem to English literature

http://www.uky.edu/~kiernan/eBeowulf/guide.ht

 

Beowulf, the community high performance computing platform

 

<Explain briefly what Beowulfs are and liken them to opensource software.>

http://www.beowulf.org/

http://www.beowulf-underground.org/

 

Objectives of this course

At the end of this course you should be able to:

(1) assemble a Beowulf cluster from a collection of PCs

(2) administer Linux systems in general and Beowulf systems in particular

(3) write programs which use the MPI library

(4) design and implement algorithms which use the message passing paradigm

 

Prerequisites for this course

Graduate status in a science or engineering program

Experience in writing programs in the C programming language

 

Course location details

 

Place for lectures

302 CERC/CRRB (on Medical campus, behind Sheetz)

Time for lectures

4-7pm Mondays

Open laboratory

711 ESB

 

Instructor details

 

Name

Frances Van Scoy

Office

G11-B CERC/CRRB

Telephone

304-293-LANE x4181

E-mail address

Fvanscoy@wvu.edu

Home page

http://www.csee.wvu.edu/~vanscoy

Office hours or by appointment

Monday

3-4

G11-B CERC

Tuesday

11-12

student lounge, 7th floor ESB

Thursday

11-12

 

Course materials

 

Text book

Thomas L. Sterling, John Salmon, Donald J. Becker, Daniel F. Savarese

How to Build a Beowulf: A Guide to the Implementation and Application of PC Clusters

MIT Press, 1999

On-line references

< MPI tutorial from PSC >

PSC Unix overview

http://www.psc.edu/general/unix/unixoverview.html

PSC Parallel Programming Techniques

http://www.psc.edu/training/PSU/Contents.html

A. PSC Environment http://www.psc.edu/training/PSU/PSC_Environment/psc_environ.html

B. Parallel Computing: Overview http://www.psc.edu/training/PSU/PC_Overview/PC_Overview.html

C. Compiling and Job Submission http://www.psc.edu/training/PSU/Compiling_Job_Submission/Comp_Job.html

D. Message Passing Basics http://www.psc.edu/training/PSU/Message_Passing/MPI_Basics.html

E. Message Passing: Code Development http://www.psc.edu/training/PSU/Code_Development/Code_Development.html

F. Performance Optimization http://www.psc.edu/training/PSU/Performance_Optimization/Performance_Optimization.html

G. TotalView Debugger http://www.psc.edu/training/PSU/Totalview/TotalView.html

Grade components

 

Programming assignments @ 7%

2 emphasizing MPI features

3 emphasizing algorithm development

35%

Term project

20%

Tests @ 10%

2 in-class hour tests

1 practical test

30%

Final exam

15%

 

Tentative calendar

 

week

date

topic

reading assignment

other assignment

1

Jan 14

     

(2)

Jan 21

(no class-MLK Day)

   

3

Jan 28

     

4

Feb 4

     

5

Feb 11

     

6

Feb 18

     

7

Feb 25

     

8

Mar 4

     

9

Mar 11

     

10

Mar 18

     

(11)

Mar 25

(no class- spring break)

   

12

Apr 1

     

13

Apr 8

     

14

Apr 15

     

15

Apr 22

     

16

Apr 29

     

 

 

 

ÿ