The Community for Technology Leaders
Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques (2005)
St. Louis, Missouri
Sept. 17, 2005 to Sept. 21, 2005
ISSN: 1089-795X
ISBN: 0-7695-2429-X
pp: 279-290
Parry Husbands , Computational Research Division, Lawrence Berkeley National Laboratory
Paul Hargrove , Computational Research Division, Lawrence Berkeley National Laboratory
Costin Iancu , Computational Research Division, Lawrence Berkeley National Laboratory
ABSTRACT
<p>Hiding communication latency is an important optimization for parallel programs. Programmers or compilers achieve this by using non-blocking communication primitives and overlapping communication with computation or other communication operations. Using non-blocking communication raises two issues: performance and programmability. In terms of performance, optimizers need to find a good communication schedule and are sometimes constrained by lack of full application knowledge. In terms of programmability, efficiently managing nonblocking communication can prove cumbersome for complex applications. In this paper we present the design principles of HUNT, a runtime system designed to search and exploit some of the available overlap present at execution time in UPC programs. Using virtual memory support, our runtime implements demand-driven synchronization for data involved in communication operations. It also employs message decomposition and scheduling heuristics to transparently improve the non-blocking behavior of applications. We provide a user level implementation of HUNT on a variety of modern high performance computing systems. Results indicate that our approach is successful in finding some of the overlap available at execution time. While system and application characteristics influence performance, perhaps the determining factor is the time taken by the CPU to execute a signal handler. Demand driven synchronization at execution time eliminates the need for the explicit management of non-blocking communication. Besides increasing programmer productivity, this feature also simpli- fies compiler analysis for communication optimizations.</p>
INDEX TERMS
null
CITATION
Parry Husbands, Paul Hargrove, Costin Iancu, "HUNTing the Overlap", Proceedings of the 22nd International Conference on Parallel Architectures and Compilation Techniques, vol. 00, no. , pp. 279-290, 2005, doi:10.1109/PACT.2005.25
97 ms
(Ver 3.3 (11022016))