NSF-CBMS Short Course on Parallel Numerical Linear Algebra
Local information for NSF-CBMS Short Course
Survey Results of Students in NSF-CBMS Short Course
Sharks and Fish
Handouts for UC Berkeley CS267 Course
Handout 1: Class Introduction for Spring 1995
Handout 2: Class Survey for Spring 1995
CS267 Spring 1994 Midterm
Assignment 1: Fast Matrix Multiply
Notes for Discussion 1: Using the CM-5, fast 2x2 matrix multiply
Assignment 2 and 3: Sharks and Fish 1 & 2
Notes for Discussion 2: The Speedup Question
Assignment 4: Particle Simulation
Comments on Assignment 2 and 3: Suggestions for Assn 4
Assignment 5: Parallel Traveling Salesman
Comments on Assignment 4
Assignment 5 details
Final Project Suggestions
Lecture 1, 1/17: Introduction to Parallel Computing
Lecture 2, 1/19: Exploiting Memory-hierarchies in Linear Algebra Kernels
Lecture 3, 1/24: The IBM RS6000/590 - architecture and algorithms.
Lecture 4, 1/26: Measuring performance of Parallel Programs.
Lecture 5, 1/31: Introduction to Parallel Programming Languages.
Lecture 6, 2/2: Message Passing Programming with CMMD.
Lecture 6(suppl), 2/2: Shared Memory Programming with Multithreading.
Lecture 7, 2/7: Parallel Programming with Split-C I.
Lecture 8, 2/9: Parallel Programming with Split-C II.
Lecture 9, 2/14: Parallel Architectures I.
Lecture 10, 2/16: Parallel Architectures II.
Lecture 11, 2/21: Mapping Matrix Multiplication to Different Architectures
Lecture 12, 2/23: LAPACK and ScaLAPACK I; Using the WWW
Lecture 13, 2/28: LAPACK and ScaLAPACK II
Lecture 13(cont), 2/28: LAPACK and ScaLAPACK II
Dense Least Squares Problems
Dense Symmetric Eigenvalue Problems
Dense Nonsymmetric Eigenvalue Problems
Ming Gu's home page
for detailed error analyses.
Lecture 14, 3/2: PRAM and Parallel Prefix.
Lecture 15, 3/7: The Multipol Library.
Lecture 16, 3/9: Sources of Parallelism and Locality in Simulation I.
Lecture 17, 3/14: Sources of Parallelism and Locality in Simulation II.
Lecture 18, 3/16: Introduction to Graph Partitioning.
Lecture 19, 3/21: Partitioning Graphs without Coordinate Information I.
Lecture 20, 3/24: Partitioning Graphs without Coordinate Information II.
Lecture 21, Floating Point, Error Analysis, Numerical Stability
Lecture 21(supplemental), 4/4: IEEE Floating Point Arithmetic
Lecture 22, 4/6: Parallel Sorting - Algorithms and Performance Analysis
Lecture 23, 4/11:Graph Partitioning (continued)
Lecture 24, 4/13: Solving the Discrete Poisson Equation using Jacobi, SOR and the FFT
Lecture 25, 4/18: Solving the Discrete Poisson Equation using Multigrid and a comparison of methods
Lecture 26, 4/20: Fast Hierarchical Methods for the N-body Problem
Lecture 27, 4/25: Fast Hierarchical Methods for the N-body Problem(continued)
Lecture 28, 4/27: Parallelizing a Global Climate Model
Lecture 29, 5/2: Parallelizing Compilers
. See also
Compilation for Parallel Machines
for the course
CS 264, Implementation of Programming Languages
Axel's draft Lanczos template
Final Projects: Spring 1995
Fast Parallel Iterative Matrix Diagonalization
Ptolemy C Code Generation and Scheduling for the Network of Workstations (NOW)
Corresponding and Re-rendering Stereo Photographs on a Shared-Memory Architecture
Parallel Raytracing using a Network of Workstations for Rendering Spline Surface Animation
Parallel Monte Carlo Simulation
Berkeley Search Engine
Porting and Characterization of GATOR, an Atmospheric Chemical Tracer Model
A Distributed Memory Concurrent B-tree Implementation
Design, Implementation, and Performance Evaluation of a Portable Distributed Task Queue
Porting The BLACS From MPL To GAM On The SP-1
Implementation of a Parallel Preconditioned Conjugate Gradient (PCG) Solver in Finite Element Analysis Program (FEAP)
Parallelizing Impulse, a dynamic simulation system.
Model of LPARX multigrid performance on the CM5
Online Documentation and Information about Machines
(Networks of Workstations)
IBM SP1 and SP2
SGI Power Challenge
Catalog of Commercial Hardware and Software Vendors
Matlab Primer, 3rd Edition
Matlab Online Reference Docs
Reference Material/Online Course Reader
Parallel Information Sites
NSF HPCC Science Highlights -- Overview
NHSE - HPCC National HPCC Software Exchange
CSEP - Computational Science Education Project
PETSc: Portable, Extensible, Toolkit for Scientific Computation
Supercomputing & Parallel Computing: Conferences & Journals
CMU's list of supercomputing and parallel computing resources
GAMS - Guide to Available Math Software
Netlib Repository at UTK/ORNL
J. Wang's Parallel Computing List.
The Parallel Computing Archive
at the University of Kent at Canterbury, U.K.
Major Science Research Institutions from Caltech
CSC Mathematical Topics
Textbooks about parallelism.
Designing and Building Parallel Programs
. A dead-tree version of this book is available by Addison-Wesley.
Parallel Related Sites at Berkeley
Castle, a portable parallel programming environment
Split-C, a performance programming language for parallel machines
Mantis, a parallel debugger for Split-C
Multipol, A data structure library for distributed memory multiprocessors
Titanium, compiler optimizations for explicitly parallel programs
NOW, a Network Of Workstations
pSather, a parallel object oriented programming language
Sequoia 2000 Global Change Researchh Project
On-Web Courses in High Performance Computing
, Parallel Architecture, Spring 1995
, Parallel Scientific Computing, Spring 1995.
, Multiprocessor Networks, Fall 1994
Other UCB EECS classes with home pages
This page has been accessed
times since Jan 16, 1995