ECS289K Homepage, Fall 2003

Instructor:: Zhaojun Bai, 3005 Engineering II, 752-4874, bai@cs.ucdavis.edu
Lecture:: Tu./Th., 1:40pm - 3:00pm, 117 Olson
Office Hours and Discussions:: Tu./Th., 11:00am - 12:30pm, or by appointment
Prerequisite: ECS130 or consent of instructor
Catalog Description: Types of parallel computing models. Message passing programming. Fundamental algorithms for sorting and selection, linear algebra, etc. Study of exiting parallel programming languages and compilers, environments, libraries and toolboxes.
Goals of the Course

Essential numerical algorithms in parallel scientific computing
Implementation of parallel algorithms on parallel cluster computers
Algorithm and program development of medium sized applications
The trends in computational science and engineering
Syllabus

Introduction
Overview of parallel computing models
Message passing programming
Primitives (parallel prefix, sorting and selection, ...)
Vector and matrix operations
Matrix algorithms
Sparse vector and matrix operations
Preconditioned conjugate gradient method
Domain decomposition
Selected topics on medium size applications
References
Grading:

Homework, computing project and presentation: 70%
Final project and report: 30%
On-line Info:: Class annoucements and handouts will be available at this site:
http://www.cs.ucdavis.edu/~bai/ECS289K/

Approximate Lecture/Meeting Schedule

9/25 Course annoucement, and introduction to parallel computing
9/30 Uniprocessor optimization,
10/2 Introduction to parallel architectures and programming models. (reading: chap 3 of the source book).
10/7 Message passing programming (MPI)
10/9 Dense matrix algorithms I
10/28 Dense matrix algorithms II
10/30 Primitives I
11/4 Primitives II
11/6 Sparse matrix computing I
11/13 Sparse matrix computing II
11/17, 6:30pm-8:30pm, Room 3085, Kemper Hall
6:30-6:55: Amitabha Banerjee , AWO
6:55-7:20: Jenny Wang, PP
7:20-7:45: Clark Crawford, RA
7:45-8:10: Ichitaro Yamazaki, SUMMA
8:10-8:35: Guanzhou Jie, SMV
11/18
1:40-2:05: Vicentiu Neagoe , AWO
2:05-2:30: Weiwei Cao , RA
2:30-2:55: Deanna Rogers, PP
11/20
1:40-2:05: Ethan Chen, AWO
2:05-2:30: Jaya Sreevalsan Nair, SMV
2:30-2:55: James Chen, SUMMA
12/2 Graph partition I
12/4 Graph partition II
12/11 Final project presentation (Room 1066, Kemper Hall)
1:00 - 1:20: Deanna Rogers/Vicentiu Neagoe
1:20 - 1:40: Guanzhou Jie
1:40 - 2:00: Amitabha Banerjee
2:00 - 2:20: Clark Crawford
2:20 - 2:40: Weiwei Cao/Ethan Chen
3:00 - 3:20: Ichitaro Yamazaki
3:20 - 3:40: Jenny Wang
3:40 - 4:00: Jaya Sreevalsan Nair
4:00 - 4:20: James Chen
12/12, 5pm, final project due

Assignments (reading, presentation, homework, project)

9/30 reading assignment: papers in ``uniprocessor performance optimization and benchmarking'' section of the reference list
9/30 project A: Optimize/tune sequential matrix multiply
10/7 reading assignment: a long tutorial on MPI by Gropp.
10/7 project B: simple MPI exercises
10/31 Reading and presentation assignment
11/13 project C: an MPI implementation of the CG method
12/11 Final project presentation
12/12 5:00pm, final project due

Maintained by Zhaojun Bai, bai@cs.ucdavis.edu.