Title | ||
---|---|---|
Communication and memory requirements as the basis for mapping task and data parallel programs |
Abstract | ||
---|---|---|
For a wide variety of applications, both task and data parallelism must be exploited to achieve the best possible performance on a multicomputer. Recent research has underlined the importance of exploiting task and data parallelism in a single compiler framework, and such a compiler can map a single source program in many different ways onto a parallel machine. The tradeoffs between task and data parallelism are complex and depend on the characteristics of the program to be executed, most significantly the memory and communication requirements, and the performance parameters of the target parallel machine. In this paper, we present a framework to isolate and examine the specific characteristics of programs that determine the performance for different mappings. Our focus is on applications that process a stream of input, and whose computation structure is fairly static and predictable. We describe three such applications that were developed with our compiler: fast Fourier transforms, narrowband tracking radar, and multibaseline stereo. We examine the tradeoffs between various mappings for them and show how the framework is used to obtain efficient mappings. |
Year | DOI | Venue |
---|---|---|
1994 | 10.1145/602770.602832 | SC |
Keywords | Field | DocType |
different way,different mapping,memory requirements,target parallel machine,data parallelism,communication requirement,mapping task,data parallel programs,single compiler framework,single source program,complex tradeoffs,memory requirement,data parallel program,parallel machine,performance parameter,possible performance,requirements,compiler,application software,computer science,parallel processing,concurrent computing,parallel programming,digital signal processing,task parallelism,fast fourier transform,load balancing,fast fourier transforms,image processing,fibre channel,computer programming,compilers | Instruction-level parallelism,Computer science,Task parallelism,Parallel computing,Compiler,Fibre Channel,Data parallelism,Fast Fourier transform,Computer programming,Computation | Conference |
ISSN | ISBN | Citations |
1063-9535 | 0-8186-6605-6 | 19 |
PageRank | References | Authors |
3.08 | 15 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jaspal Subhlok | 1 | 671 | 76.68 |
David R. O'hallaron | 2 | 1243 | 126.28 |
Thomas Gross | 3 | 19 | 3.08 |
Peter A. Dinda | 4 | 1493 | 126.40 |
Jon Webb | 5 | 21 | 4.26 |