A recursive doubling algorithm for solution of tridiagonal systems on hypercube multiprocessors (Q1822894): Difference between revisions
From MaRDI portal
Changed an Item |
Changed an Item |
||
| Property / describes a project that uses | |||
| Property / describes a project that uses: LINPACK / rank | |||
Normal rank | |||
Revision as of 05:33, 28 February 2024
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | A recursive doubling algorithm for solution of tridiagonal systems on hypercube multiprocessors |
scientific article |
Statements
A recursive doubling algorithm for solution of tridiagonal systems on hypercube multiprocessors (English)
0 references
1989
0 references
The authors give a limited processor version of the recursive doubling algorithm for the solution of tridiagonal linear systems using parallel arithmetic steps on a parallel computer. They make use of fast parallel prefix algorithms. The algorithm achieves linear speedup and constant efficiency over its sequential implementation. Computational results are given based on an Intel iPSC/d5 hypercube multiprocessor. The authors are in the process of extending the proposed algorithm to general recurrence relations, block tridiagonal and banded linear systems.
0 references
block tridiagonal linear systems
0 references
recursive doubling algorithm
0 references
parallel computer
0 references
parallel prefix algorithms
0 references
linear speedup
0 references
efficiency
0 references
hypercube multiprocessor
0 references
recurrence relations
0 references
banded linear systems
0 references
