Parallel and Distributed Computation (CS621, Fall 2010)

Instructor: Jun Zhang, E-mail:jzhang@cs.uky.edu, Tel:257-3892.
Office: 763F Anderson Tower.
Office Hours: WF: 1:00 - 2:00pm, and by appointment.

Text Book:
Introduction to Parallel Computing: Design and Analysis of Algorithms
by Vipin Kuman, Ananth Grama, Anshul Gupta, and George Karypis, 2nd Ed., 2003. (required, BOOK 1)

Using MPI: Portable Parallel Programming with the Message-Passing Interface
by William Gropp, Ewing Lusk, and Anthony Skjellum, 2nd Ed., 1999. (BOOK 2)

CS621 syllabus.ps, or syllabus.pdf, for Fall 2010; (subject to change)

Some rules rule.ps, or rule.pdf, you should know before taking this course.

Here is An MPI Tutorial, describing simple MPI routines

• First assignment: Read through Chapter 1 of both BOOK 1 and BOOK 2.
• Project 1, assigned on 9/7, due on September 14, 2004;
• Homework 1, ps file, and pdf file, due September 28;
• Project 2, ps file, and pef file, due October 19;
• Homework 2, ps file, and pdf file, due Novemver 16;
• Project 3, ps file, and pdf file, due December 2;
• The 2001 Midterm Exam Problems, ps file, and pdf file.
  The 1999 Midterm Exam Problems, ps file, and pdf file, and

  These past exams are only provided for reference purpose. There is no guarantee that this year's midterm exam will be similar to those ones;

For some informtion on building PC and workstation clusters, please see Microway, Inc.

Message Passing Interface: We plan to use the emerging standard Message Passing Interface (MPI) on the Power Challenge. Although this is not a very efficient way to do parallel computing on Power Challenge, our goal is to learn programming skill with MPI. Here is the official MPI site.

Here is a description of the course. I am working on a postscript file of the course description. It is called syllabus.ps. You should also read the general policies on homework and grading, here is the downloadable rule.ps.

• September 2; September 4; September 8; September 14; September 16; September 18;
• September 21; September 23; September 25; September 28; September 30; October 5;
• October 7; October 8; October 9; October 12; October 14; October 16; October 26;
• October 30; November 1; November 4; November 6; November 9; November 11;
• November 13; November 16; November 18; November 20; November 23; November 25;
• December 2; December 4; December 7; December 9;

• Chapter 1; August 30
• September 9; September 14;
October 7; October 14; October 26; November 30; December 2; December 7;

Please be advised that the midterm exam is on Tuesday October 22.

• Set up your HP Supercomputer Account.
  If you decided to take this course, please e-mail me with information on your name and student ID number.
• To know how to run MPI on the the HP supercomputer, see Project0.html.
• You need to download Project0.tar. These are sample makefile and C and Fortran codes.

• IEEE Computer Society's ParaScope, a list of parallel computing sites.
• On February, 1998, IBM announced the world's first 1000 MHZ chip, three times faster than Intel's fastest chip. Don't worry about that, your parallel computers could be even faster.
• In May, 1998, Los Alamos built a mail-order supercomputer and it is among the word's fastest. Take a look at the new supercomputer, named Avalon.
• The TOP500 List of Supercomputer Sites .
• MPI Software Technology, Inc.: a commercial MPI software company.
• Beowulf Project at CESDIS.
• Take a look at the Earth Simulator in Japan, the fastest supercomputer on Earth.

• CRAY SV1.
  
• CRAY T3E.
  
• CRAY J90.
  
• CRAY ORIGIN2000.
  

• University of Minnesota.
• Pennsylvania State University.
• Stanford University
• University of California at Berkeley.
• University of California at Berkeley.
• University of California at Berkeley.
• University of Illinois at Urbana-Champaign.
• University of Tennessee at Knoxville.
• University of Tennessee at Knoxville.

• Designing and Building Parallel Programs by Ian Foster.

Return to Jun Zhang's:
Home page; Publication list; Research interests; Resume.