Professor: Peter Pacheco
Office: Harney 540
Phone: 422-6630
Email: user: peter, domain: usfca.edu
Office Hours: MW 5:30-7:00 and by appointment

Course Syllabus

Midterm, which is due Wednesday, March 12 at 4:45 pm. Here's a key.

Programming Assignments

1. Programming Assignment 1. Due at 3pm on Monday, February 11.
2. Programming Assignment 2. Note that the due date has been changed to Monday, March 10.
3. Programming Assignment 3. Due at 3pm on Monday, April 7.

Seminar Papers

• The first three seminars will discuss "The Landscape of Parallel Computing Research: A View From Berkeley," by Krste Asanovic, et al.
  1. First presentation: Introduction, Motivation, Applications and Dwarfs, pp. 1-19. Garrik Sturges will lead the discussion.
  2. Second presentation: Hardware and Programming Models Dwarfs, pp. 20-36. Bina Maniar will lead the discussion.
  3. Third presentation: Systems Software, Metrics for Success, and Conclusion, pp. 36-45. Shaoting Cai will lead the discussion.
• The fourth seminar will discuss, ``Transactional Memory, An Overview,'' by Tim Harris, et al. Doug Campbell will lead the discussion.
• The fifth seminar will discuss ``Threads cannot be implemented as a library'' (link accessible from CS department machines) by Hans-J. Boehm. Nicholas Ilacqua will lead the discussion.
• The sixth seminar will discuss ``MapReduce: simplified data processing on large clusters'' (link accessible from CS department machines) by Jeffrey Dean and Sanjay Ghemawat. Xi Zhang (Xeno) will lead the discussion.
• The seventh seminar will discuss ``Python-Based Distributed Programming with Trickle'' by G.D. Benson and A.S. Fedosov. Peng Liu and Phimphan Yoophoon will lead the discussion.
• The eighth and ninth seminars will discuss ``Parallel Programmability and the Chapel Language'' by Bradford L. Chamberlain, et al.
  1. First presentation: Sections 1, 2, and 3 (pages 1-12). Sunly Guo will lead the discussion.
  2. Second presentation: Sections 4 and 5 (pages 12-21). Ted Johnson will lead the second discussion.
• The ninth seminar will discuss ``Optimization of Sparse Matrix-Vector Multiplication on Emerging Multicore Platforms'' by S. Williams, et al. Xue Yin will lead lead the discussion.
• The tenth seminar will discuss ``Scientific Application Performance on Candidate PetaScale Platforms'' by L. Oliker, et al. Suli Xu will lead the discussion.
• The eleventh and last seminar will discuss ``High Performance Development for High End Computing With Python Language Wrapper (PLW)'' by Piotr Luszczek and Jack Dongarra. Lili Sun will lead the discussion.

Additional Course Information

• Paper Evaluation Form
• Paper Evaluation Form (plain text)
• Setting Up a Subversion Repository for CS 625
• Using the Penguin Cluster
• Predicted speedups and efficiencies of block-row partitioned matrix-vector multiplication
• A list of possible projects for the last programming assignment

Code

• Implementations of the Trapezoidal Rule
  1. A serial implementation
  2. An MPI implementation
  3. A Pthreads implementation
  4. First OpenMP implementation
  5. Second OpenMP implementation
  6. Third OpenMP implementation
• A macro for taking elapsed times