Partitioning and scheduling workflows across multiple sites with storage constraints

Authors:
Weiwei Chen;Ewa Deelman
Affiliations:
Information Sciences Insitute, University of Southern California, Marina del Rey, CA;Information Sciences Insitute, University of Southern California, Marina del Rey, CA
Venue:
PPAM'11 Proceedings of the 9th international conference on Parallel Processing and Applied Mathematics - Volume Part II
Year:
2011

Citing 7
Cited 1

Performance-Effective and Low-Complexity Task Scheduling for Heterogeneous Computing

IEEE Transactions on Parallel and Distributed Systems
Scheduling of scientific workflows in the ASKALON grid environment

ACM SIGMOD Record
Task scheduling strategies for workflow-based applications in grids

CCGRID '05 Proceedings of the Fifth IEEE International Symposium on Cluster Computing and the Grid (CCGrid'05) - Volume 2 - Volume 02
Workflows for e-Science: Scientific Workflows for Grids

Workflows for e-Science: Scientific Workflows for Grids
Pegasus: A framework for mapping complex scientific workflows onto distributed systems

Scientific Programming
Optimizing workflow data footprint

Scientific Programming - Dynamic Computational Workflows: Discovery, Optimization and Scheduling
Run-time Optimisation of Grid Workflow Applications

GRID '06 Proceedings of the 7th IEEE/ACM International Conference on Grid Computing

Integration of Workflow Partitioning and Resource Provisioning

CCGRID '12 Proceedings of the 2012 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (ccgrid 2012)

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper aims to address the problem of scheduling large workflows onto multiple execution sites with storage constraints. Three heuristics are proposed to first partition the workflow into sub-workflows. Three estimators and two schedulers are then used to schedule sub-workflows to the execution sites. Performance with three real-world workflows shows that this approach is able to satisfy storage constraints and improve the overall runtime by up to 48% over a default whole-workflow scheduling.