Scheduler oriented grid performance evaluation

Authors:
Liang Peng;Simon See
Affiliations:
Asia Pacific Science and Technology Center, Sun Microsystems Inc. Nanyang Center for Supercomputing and Visualization, Nanyang Technological University, Singapore;Asia Pacific Science and Technology Center, Sun Microsystems Inc. Nanyang Center for Supercomputing and Visualization, Nanyang Technological University, Singapore
Venue:
PDCAT'04 Proceedings of the 5th international conference on Parallel and Distributed Computing: applications and Technologies
Year:
2004

Citing 3
Cited 0

The network weather service: a distributed resource performance forecasting service for metacomputing

Future Generation Computer Systems - Special issue on metacomputing
Dynamic Performance Forcasting for Network-Enabled Servers in a Metacomputing Environment

IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
Tools and Techniques for Measuring and Improving Grid Performance

IWDC '02 Proceedings of the 4th International Workshop on Distributed Computing, Mobile and Wireless Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Grid computing is a fast developing technology as an approach to do high performance scientific and engineering computation. The research work on Grid performance evaluation and modeling, an important supporting factor, is still immature and not considered inherently grid middleware design. In this paper, we proposed a scheduler oriented grid performance evaluation approach so that it closely cooperates with grid schedulers. A typical scenario of how the the grid performance modules interact with the scheduler is also introduced and analyzed. The advantage is that the grid scheduler is able to get more performance information about remote sites and hence improve the scheduling policy for grid jobs and ensure the quality of service to some extent.