Job-scheduling via resource availability prediction for volunteer computational grids

Authors:
Jun Zhang;Chris Phillips
Affiliations:
School of Electronic Engineering and Computer Science, Queen Mary, University of London, Mile End Road, E1 4NS, London, UK.;School of Electronic Engineering and Computer Science, Queen Mary, University of London, Mile End Road, E1 4NS, London, UK
Venue:
International Journal of Grid and Utility Computing
Year:
2011

Citing 18
Cited 1

The interaction of parallel and sequential workloads on a network of workstations

Proceedings of the 1995 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
Heuristic Algorithms for Scheduling Independent Tasks on Nonidentical Processors

Journal of the ACM (JACM)
The network weather service: a distributed resource performance forecasting service for metacomputing

Future Generation Computer Systems - Special issue on metacomputing
A comparison of eleven static heuristics for mapping a class of independent tasks onto heterogeneous distributed computing systems

Journal of Parallel and Distributed Computing
Matching and Scheduling Algorithms for Minimizing Execution Time and Failure Probability of Applications in Heterogeneous Computing

IEEE Transactions on Parallel and Distributed Systems
Dynamically forecasting network performance using the Network Weather Service

Cluster Computing
The ANL/IBM SP Scheduling System

IPPS '95 Proceedings of the Workshop on Job Scheduling Strategies for Parallel Processing
Scheduling Resources in Multi-User, Heterogeneous, Computing Environments with SmartNet

HCW '98 Proceedings of the Seventh Heterogeneous Computing Workshop
The Relative Performance of Various Mapping Algorithms is Independent of Sizable Variances in Run-time Predictions

HCW '98 Proceedings of the Seventh Heterogeneous Computing Workshop
A Comparison among Grid Scheduling Algorithms for Independent Coarse-Grained Tasks

SAINT-W '04 Proceedings of the 2004 Symposium on Applications and the Internet-Workshops (SAINT 2004 Workshops)
Predicting node availability in peer-to-peer networks

SIGMETRICS '05 Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Distributed computing in practice: the Condor experience: Research Articles

Concurrency and Computation: Practice & Experience - Grid Performance
Scheduling task parallel applications for rapid turnaround on desktop grids

Scheduling task parallel applications for rapid turnaround on desktop grids
Improving distributed system performance using machine availability prediction

ACM SIGMETRICS Performance Evaluation Review
Characterizing resource availability in enterprise desktop grids

Future Generation Computer Systems
Resource Availability Prediction for Improved Grid Scheduling

ESCIENCE '08 Proceedings of the 2008 Fourth IEEE International Conference on eScience
Multi-state grid resource availability characterization

GRID '07 Proceedings of the 8th IEEE/ACM International Conference on Grid Computing
Modeling machine availability in enterprise and wide-area distributed computing environments

Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing

OXDP & OXiP: the notion of objects for efficient large XML data queries

International Journal of Grid and Utility Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In a volunteer-based computational grid computing, one big challenge for effective job allocation is resource availability. As resources in this environment are volatile, matching guest jobs to suitable resources is very important. To improve scheduling, especially in terms of avoiding job failures due to resource unavailability, we propose a new job-scheduling algorithm called first-come-first-served plus predictor (FCFSPP). This scheduling algorithm is based on an existing resource availability prediction method that anticipates the future availability of resources to help make reliable job allocation decisions. According to the simulation results, FCFSPP does not only reduce the number of job failures but also maintain acceptable job throughput in volatile volunteer environments by providing reliable job allocation decisions.