Given by Geoffrey C. Fox at CPS615 Basic Simulation Track for Computational Science on Fall Semester 95. Foils prepared 29 August 1995
Outside Index
Summary of Material
Overview of National Program -- The Grand Challenges |
Overview of Technology Trends leading to petaflop performance in year 2015 |
Overview of Syracuse and National programs in computational science |
Parallel Computing in Society |
Parallel and Sequential Computer Architectures |
Why Parallel Computing works |
Message Passing and Data Parallel Programming Paradigms |
Laplace Equation with Iterative solver in detail |
Set (approximately 6) of application/algorithm snippets illustrating software, hardware and algorithm issues |
Outside Index Summary of Material
Geoffrey Fox |
NPAC |
Room 3-131 CST |
111 College Place |
Syracuse NY 13244-4100 |
Overview of National Program -- The Grand Challenges |
Overview of Technology Trends leading to petaflop performance in year 2015 |
Overview of Syracuse and National programs in computational science |
Parallel Computing in Society |
Parallel and Sequential Computer Architectures |
Why Parallel Computing works |
Message Passing and Data Parallel Programming Paradigms |
Laplace Equation with Iterative solver in detail |
Set (approximately 6) of application/algorithm snippets illustrating software, hardware and algorithm issues |
Instructor: Geoffrey Fox gcf@npac.syr.edu 3154432163 Room 3-131 CST |
Backup: Nancy McCracken njm@npac.syr.edu 3154434687 Room 3-234 CST |
TA: John Houle houle@npac.syr.edu |
NPAC Administrative support: Nora Downey-Easter nora@npac.syr.edu 3154434740 Room 3-210 CST |
CPS615 Powers that be above can be reached at cps615ad@npac.syr.edu |
CPS615 Students can be reached by mailing cps615@npac.syr.edu |
Homepage will be: |
http://www.npac.syr.edu/projects/cps615fall95 |
Graded on the basis of Approximately 8 Homeworks which will be due Wednesday of week following day given out (Monday or Wednesday) |
Plus one small project at the end of class |
No finals or written exams |
All material will be placed on World Wide Web(WWW) |
Preference given to work returned on the Web -- optional lecture will be given on how to use WWW |
Overview of National Scene -- Why is High Performance Computing Important |
What is Computational Science -- The Program at Syracuse |
Basic Technology Situation -- Increasing density of transistors on a chip |
Elementary Discussion of Parallel Computing |
Computer Architecture -- Parallel and Sequential
|
Simple base example -- Laplace's Equation |
Programming Models -- Message Passing and Data Parallel Computing -- MPI and HPF
|
This introduction is followed by a set of "vignettes" discussing applications and algorithms which illustrate parallel programming and parallel algorithms
|
There are several machines still being used but the number of vendors and diversity of designs has shrunk |
The SIMD Maspar and AMT DAP are focusing on database and other niche markets such as signal processing. They are not considered mainstream any longer. |
The MIMD nCUBE3 is not deployed and the nCUBE2 is uncompetitive in science arena. The company is focusing on video server market. |
The major pure MIMD distributed memory machines are the IBM SP-2 and Intel Paragon with the IBM having the best node CPU and the Paragon a superior network |
There is a spectrum of shared memory machines from
|