Proceedings. 1998 International Conference on Parallel Architectures and Compilation Techniques (Cat. No.98EX192) (1998)
Oct. 12, 1998 to Oct. 18, 1998
Luc Bougé , LIP, ENS Lyon
Raymond Namyst , LIP, ENS Lyon
Christian Perez , LIP, ENS Lyon
Phil Hatcher , University of New Hampshire
This paper studies the benefits of compiling data-parallel languages onto a multithreaded runtime environment providing dynamic thread migration facilities. Each abstract process is mapped onto a thread, so that dynamic load balancing can be achieved by migrating threads among the processing nodes. We describe and evaluate an implementation of this idea in the Adaptor HPF compiler. We show that no deep modification of the compiler are needed, and that the overhead of managing threads can be kept small. As an experimental validation, we report on an HPF implementation of the Gauss Partial Pivoting algorithm. We show that using an initial BLOCK data distribution with our dynamic load balancing scheme can reach the performance of the optimal CYCLIC distribution.
Data-parallel languages, Compilation, Load Balancing, Thread migration.
L. Bougé, C. Perez, P. Hatcher and R. Namyst, "A Multithreaded Runtime Environment with Thread Migration for A HPF Data-Parallel Compiler," Proceedings. 1998 International Conference on Parallel Architectures and Compilation Techniques (Cat. No.98EX192)(PACT), Paris, France, 1998, pp. 418.