Skip to Main content Skip to Navigation

# Bi-objective Scheduling Algorithms for Optimizing Makespan and Reliability on Heterogeneous Systems

2 ALGORILLE - Algorithms for the Grid
INRIA Lorraine, LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : We tackle the problem of scheduling task graphs onto a heterogeneous set of machines, where each processor has a probability of failure governed by an exponential law. The goal is to design algorithms that optimize both makespan and reliability. First, we provide an optimal scheduling algorithm for independent unitary tasks where the objective is to maximize the reliability subject to makespan minimization. For the bi-criteria case we provide an algorithm that approximates the Pareto-curve. Next, for independent non-unitary tasks we show that the product {\em\{failure rate\}}$\times$ {\em\{unitary instruction execution time\}} is crucial to distinguish processors in this context. Based on this results we are able to let the user choose a trade-off between reliability maximization and makespan minimization. For general task graph we provide a method for converting scheduling heuristics on heterogeneous cluster into heuristics that take reliability into account. Here again, we show how we can help the user to choose a trade-off between makespan and reliability.
keyword :
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-00155964
Contributor : Erik Saule <>
Submitted on : Tuesday, June 19, 2007 - 4:24:48 PM
Last modification on : Thursday, November 19, 2020 - 12:59:40 PM
Long-term archiving on: : Friday, September 21, 2012 - 4:40:28 PM

### File

spaa078-jeannot.pdf
Publisher files allowed on an open archive

### Citation

Jack Dongarra, Emmanuel Jeannot, Erik Saule, Zhiao Shi. Bi-objective Scheduling Algorithms for Optimizing Makespan and Reliability on Heterogeneous Systems. 19th ACM Symposium on Parallelism in Algorithms and Architectures - SPAA'07, Jun 2007, San Diego, United States. pp.280-288, ⟨10.1145/1248377.1248423⟩. ⟨hal-00155964⟩

Record views

Files downloads