Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Mapping algorithms onto parallel architectures

Thesis/Dissertation ·
OSTI ID:6093069

The efficient implementation of an algorithm on a distributed-memory parallel architecture requires a mapping of processes onto processors that minimizes the total cost. A graph-partitioning model defines the cost of a mapping. The mapping problem, an NP-complete combinatorial optimization problem, is solved using a heuristic method known as simulated annealing. When applied to mapping algorithms onto hypercube interconnected distributed-memory multiprocessors, simulated annealing yields a near-optimal solution.

Research Organization:
New Mexico State Univ., Las Cruces, NM (USA)
OSTI ID:
6093069
Country of Publication:
United States
Language:
English

Similar Records

Mapping vision algorithms to parallel architectures
Journal Article · Mon Aug 01 00:00:00 EDT 1988 · Proc. IEEE; (United States) · OSTI ID:6525907

On partitioning of algorithms for parallel execution on VLSI circuit architectures
Thesis/Dissertation · Thu Dec 31 23:00:00 EST 1987 · OSTI ID:6930573

Heuristic approaches to task allocation for parallel computing
Thesis/Dissertation · Thu Dec 31 23:00:00 EST 1987 · OSTI ID:5921843