No label defined (Q4827372)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 2116061
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 2116061 |
Statements
16 November 2004
0 references
parallel scientific computation BSP
0 references
MPI
0 references
LU decomposition
0 references
FFT
0 references
sparse matrix-vector multiplication
0 references
fast Fourier transform
0 references
0.7863940000534058
0 references
The subtitle is ``using BSP and MPI''. However, the book is a course in parallel programming using BSP (bulk synchronous parallel) together with C. BSP is a communication method and library, based on one-sided communication, that allows an easy theoretical approach for analyzing parallel programs. Unfortunately in the register of the book lacks the notion ``latency hiding'' which is an essential element in parallel programming. In an introductory chapter the basic notation for parallel programming is introduced. In the three main chapters the LU decomposition, the fast Fourier transform and the sparse matrix multiplication are discussed in detail with all aspects of parallelization. In the Appendix C ``Programming in BSP style using MPI'' some remarks are made how to use the one-sided communication of MPI-2 to replace the BSP statements. As for complicated large parallel codes a theoretical investigation is anyway impossible, I do not see an advantage for BSP. Nevertheless the book is a good lecture to learn the basic lessons of parallel programming.
0 references
