Scheduling Strategies for Master-Slave Tasking on Heterogeneous Processor Platforms
BEAUMONT, Olivier
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
CARTER, Larry
Department of Computer Science and Engineering [Univ California San Diego] [CSE - UC San Diego]
See more >
Department of Computer Science and Engineering [Univ California San Diego] [CSE - UC San Diego]
BEAUMONT, Olivier
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
Laboratoire Bordelais de Recherche en Informatique [LaBRI]
Algorithms and high performance computing for grand challenge applications [SCALAPPLIX]
CARTER, Larry
Department of Computer Science and Engineering [Univ California San Diego] [CSE - UC San Diego]
Department of Computer Science and Engineering [Univ California San Diego] [CSE - UC San Diego]
FERRANTE, Jeanne
Department of Computer Science and Engineering [Univ California San Diego] [CSE - UC San Diego]
< Reduce
Department of Computer Science and Engineering [Univ California San Diego] [CSE - UC San Diego]
Language
en
Article de revue
This item was published in
IEEE Transactions on Parallel and Distributed Systems. 2004, vol. 15, p. 319―330
Institute of Electrical and Electronics Engineers
English Abstract
We consider the problem of allocating a large number of independent, equal-sized tasks to a heterogeneous computing platform. We use a nonoriented graph to model the platform, where resources can have different speeds of ...Read more >
We consider the problem of allocating a large number of independent, equal-sized tasks to a heterogeneous computing platform. We use a nonoriented graph to model the platform, where resources can have different speeds of computation and communication. Because the number of tasks is large, we focus on the question of determining 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). In contrast to minimizing the total execution time, which is NP-hard in most formulations, we show that finding the optimal steady state can be solved using a linear programming approach and, thus, in polynomial time. Our result holds for a quite general framework, allowing for cycles and multiple paths in the interconnection graph, and allowing for several masters. We also consider the simpler case where the platform is a tree. While this case can also be solved via linear programming, we show how to derive a closed-form formula to compute the optimal steady state, which gives rise to a bandwidth-centric scheduling strategy. The advantage of this approach is that it can directly support autonomous task scheduling based only on information local to each node; no global information is needed. Finally, we provide a theoretical comparison of the computing power of tree-based versus arbitrary platforms.Read less <
Origin
Hal imported