Dynamic Matching and Scheduling of a Class of Independent Tasks onto Heterogeneous Computing Systems
HCW '99 Proceedings of the Eighth Heterogeneous Computing Workshop
Heuristics for Scheduling Parameter Sweep Applications in Grid Environments
HCW '00 Proceedings of the 9th Heterogeneous Computing Workshop
Scientific workflow management and the Kepler system: Research Articles
Concurrency and Computation: Practice & Experience - Workflow in Grid Systems
Nimrod/K: towards massively parallel dynamic grid workflows
Proceedings of the 2008 ACM/IEEE conference on Supercomputing
Parameter Space Exploration Using Scientific Workflows
ICCS '09 Proceedings of the 9th International Conference on Computational Science: Part I
Scheduling workflow applications on processors with different capabilities
Future Generation Computer Systems - Collaborative and learning applications of grid technology
E-SCIENCE '09 Proceedings of the 2009 Fifth IEEE International Conference on e-Science
Wings: Intelligent Workflow-Based Design of Computational Experiments
IEEE Intelligent Systems
Scheduling parameter sweep workflow in the Grid based on resource competition
Future Generation Computer Systems
Hi-index | 0.00 |
Scheduling of Grid workflows has been a prevalent research area as it is the mechanism that helps improving the efficiency of the Grid workflow execution. The most common goal of scheduling workflow in the Grid is to minimize execution makespan and many algorithms have been proposed to address this issue. However, most of the algorithms so far usually focus and are implemented based on a single principle and act according to that principle. Only few algorithms consider adaptive scheduling process that acts differently in different situations. In this paper, we propose an adaptive Grid workflow scheduling that reacts to the presence of bottlenecks and different execution context. With this adaptive approach, the proposed algorithm can achieve better overall makespan. Simulations of parameter sweep workflows with single instance and multiple instances executed in parallel are used to evaluate the algorithm against four existing algorithms.