Vector and Parallel Computing: Issues in Applied Research and DevelopmentJ. J. Dongarra Vector and parallel computing is a fast expanding area of computing science, of relevance to many companies engaging in research into the commercial viability of parallel computing. This volume collates the latest research findings in this area. |
Contents
DivideandConquer Algorithms for the Computation of the | 1 |
A Pipelined QR Factorization Algorithm with | 11 |
Lattice QCD As a Large Scale Scientific Computation | 21 |
Copyright | |
24 other sections not shown
Other editions - View all
Common terms and phrases
algorithm Alliant applications architectures array benchmark BLAS block butterflies called clause cluster column communication compiler concurrency condition number conjugate gradient conjugate gradient method convergence debugging defined diagonal differential equations distributed Domain Decomposition Domain Decomposition Methods efficient eigenvalues elements example Figure flow fluid Fortran functional languages graphics grid hypercube implementation input integration interface iteration lattice linear systems LINPACK loop LU factorization machine mapping Math matrix memory latency method MFLOPS multiprocessor node number of processors obtained operations optimal OR-parallel output parallel computers parallel execution parallel programming partial differential equations performance preconditioner preconditioning predicates problem procedure Prolog proof tree routines scalar Schur complement sequence sequential shared memory SIAM side-effects simulation solution solve specific speedup structure subdomains subroutine supercomputers SUPRENUM symmetric task techniques tion transform triangular update values vector processors win-pak window