Scheduling Strategies for Mixed Data and Task Parallelism on Heterogeneous Clusters and Grids

Olivier Beaumont (1, 2) , Arnaud Legrand (3) , Yves Robert (3)

1 SCALAPPLIX - Algorithms and high performance computing for grand challenge applications
2 LaBRI - Laboratoire Bordelais de Recherche en Informatique
3 GRAAL - Algorithms and Scheduling for Distributed Heterogeneous Platforms

We consider the execution of a complex application on a heterogeneous "Grid" computing platform. The complex application consists of a suite of identical, independent problems to be solved. In turn, each problem consists of a set of tasks. There are dependences (precedence constraints) between these tasks. A typical example is the repeated execution of the same algorithm on several distinct data samples. We use a non-oriented graph to model the Grid platform, where resources have different speeds of computation and communication. We show how to determine the optimal steady-state scheduling strategy for each processor (the fraction of time spent computing and the fraction of time spent communicating with each neighbor). This result holds for a quite general framework, allowing for cycles and multiple paths in the platform graph.

Domaines: Calcul parallèle, distribué et partagé [cs.DC]

Olivier Beaumont, Arnaud Legrand, Yves Robert. Scheduling Strategies for Mixed Data and Task Parallelism on Heterogeneous Clusters and Grids. PDP'2003, 11th Euromicro Workshop on Parallel, Distributed and Network-Based Processing, Feb 2003, Gênes, Italy. pp.209―216, ⟨10.1109/EMPDP.2003.1183590⟩. ⟨hal-00789453⟩