The ScaLAPACK Project
Outline
Scalable Parallel Library for Numerical Linear Algebra
ScaLAPACK
Goals - Port LAPACK to distributed-memory environments.
Programming Style
Overall Structure of Software
PBLAS
ScaLAPACK Structure
What’s Included
Data Distributions
Distribution and Storage
Parallelism in ScaLAPACK
Heterogeneous Computing
Performance
LU Performance (Mflop/s) on 32 nodes Intel XP/S MP Paragon
Performance of LU fact. + solve on the Intel MP XP/S Paragon (Mflop/s) (2 computational processors per node)
Out of Core Software Approach
Out of Core Algorithms
HPF Version
HPF Performance for LU on 12-node Cluster Sun Ultra using PGI Compiler
ScaLAPACK - Ongoing Work
Summary
ScaLAPACK Team
Sparse Gaussian Elimination
Super LU
Email: dongarra@cs.utk.edu
Home Page: http://www.netlib.org/utk/people/JackDongarra