This page has been setup to allow people to provide feedback to everyone else in the class about some of the details of lectures, assignements, compilation, software and hardware, etc. Please contribute...wiki-away!
In orderew and use the wiki you will need to create an account (username, password, e-mail address and name) from the Main Page of the wiki (upper right corner). You can learn about the wiki syntax from a link on the Main Page of the wiki. It is essentially similar to HTML but simpler.
I have also added Math/LaTeX support in the wiki so you can feel free to use it. Image upload is also allowed.
Useful References
The following is a list of useful references (none of which are required)
for the material covered in CS238.
-
Numerical Linear Algebra for High-Performance Computers. Jack Dongarra,
Ian Duff, Danny Sorenson, and Henk van der Vort, SIAM. - One of the primary
references for the course.
-
Using MPI, 2nd ed. W. Gropp, Ewing Lusk, and A. Skjellum, MIT Press
(1999). - Another reference for the course and a good introduction to MPI.
-
MPI: The Complete Reference, 2nd ed. M. Snir, S. Otto, S. Huss-Lederman,
D. Walker, and J. Dongarra, MIT Press (1998) - In the same vein as K&R
for C programming. More a definition document for MPI than a textbook,
but an essential reference for the experienced MPI programmer.
-
Computer Architecture: A Quantitative Approach, 2nd ed. John Hennessy
and David Patterson, Morgan & Kaufmann (1996) - THE book for microprocessor
design and the tradeoffs involved in various designs. If you spend
many hours programming, you should take the time to read this book to gain
an appreciation about how your code looks to the CPU and memory systems.
For this course we would recommend reading the following chapters: Ch. 1-3
(ISA design + pipelining), Ch. 5 (Cache based memory hierarchies), Ch.
8 (Multiprocessors and a description of junior), and Appendix B
(Vector processors). This book is used in EE
282.
-
Parallel Computer Architecture: A Hardware/Software Approach. David
Culler, Jaswinder Pal Singh, with Anoop Gupta, Morgan & Kaufmann (1998)
- Has just about everything concerning parallel computing and architecture
design, but is a little too dense at times. For this course we would
recommend looking at the following chapters: Ch. 1-4 (Parallel programming
and workload-driven evaluation), Ch. 5.1-5.2, 5.5 (Cache coherence for
shared memory machines, ie. junior and Sun Enterprise machines),
Ch. 7 (Scalable Multiprocessors) good discussion of various machines out
there, Ch. 8.1-8.5 (Directory-based cache coherence (NUMA), ie. SGI Origin
2000/3000 and SGI Altix), Ch. 10.1-10.4, 10.9 (Interconnection networks),
Appendix: Parallel Benchmark Suites (NAS, TPC, ScaLAPACK, etc). This
book is used in CS315A
.
-
Domain Decomposition: Parallel Multilevel Methods for Elliptic Partial
Differential Equations. Barry Smith, Peter Bjorstad, and William Gropp,
Cambridge University Press. - A discussion of a popular way to achieve
good parallel performance from various scientific computations.
Web Links
-
MPI Links
-
OpenMP
-
Scientific Linear Algebra Libraries
- BLAS - Basic Linear Algebra Subprograms
- LAPACK - Linear Algebra PACKage
- BLACS - Basic Linear Alegbra Communication Subprograms
- PBLAS - Parallel Basic Linear Algebra Subprograms
- ScaLAPACK - Scalable LAPACK
-
ASCI/DOE Supercomputing Links
-
NSF/NPACI/DoD Supercomputing Links
-
Vendors
-
Parallel Computing Tools
-
Compendium
TA's Special Links
Last Modified: Thu Mar 31 22:05:18 PST 2005
Questions & Comments
