Execution models for mapping programs onto distributed-memory parallel computers. Final Report
The problem of exploiting the parallelism available in a program to efficiently employ the resources of the target machine is addressed. The problem is discussed in the context of building a mapping compiler for a distributed memory parallel machine. The paper describes using execution models to drive the process of mapping a program in the most efficient way onto a particular machine. Through analysis of the execution models for several mapping techniques for one class of programs, we show that the selection of the best technique for a particular program instance can make a significant difference in performance. On the other hand, the results of benchmarks from an implementation of a mapping compiler show that our execution models are accurate enough to select the best mapping technique for a given program.
- Research Organization:
- National Aeronautics and Space Administration, Hampton, VA (United States). Inst. for Computer Applications in Science and Engineering
- OSTI ID:
- 5186699
- Report Number(s):
- N-92-21251; NASA-CR-189613; NAS-1.26:189613; ICASE-92-8; CNN: NAS1-18605; MDA972-90-C-0035; ARPA ORDER 7330
- Country of Publication:
- United States
- Language:
- English
Similar Records
Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)
Consistent first solution speedups in OR-parallel execution of logic programs
Related Subjects
PARALLEL PROCESSING
ALGORITHMS
ARRAY PROCESSORS
COMPUTER ARCHITECTURE
COMPUTER CODES
DISTRIBUTED DATA PROCESSING
IMPLEMENTATION
MEMORY MANAGEMENT
PERFORMANCE
TRANSLATORS
DATA PROCESSING
MATHEMATICAL LOGIC
PROCESSING
PROGRAMMING
990200* - Mathematics & Computers