Analysis of Jobs in a Multi-Organizational Grid Test-bed

Authors:
Bu-Sung Lee;Ming Tang;Junwei Zhang;Ong Yew Soon;Cindy Zheng;Peter Arzberger;David Abramson
Affiliations:
Nanyang Technological University, Singapore;Nanyang Technological University, Singapore;Nanyang Technological University, Singapore;Nanyang Technological University, Singapore;San Diego Supercomputer Center, USA;San Diego Supercomputer Center, USA;Monash University, Australia
Venue:
CCGRID '06 Proceedings of the Sixth IEEE International Symposium on Cluster Computing and the Grid
Year:
2006

Citing 0
Cited 3

Lessons learned through driving science applications in the PRAGMA grid

International Journal of Web and Grid Services
National Grid+NRENs in ASEAN region: a snap shot

Proceedings of the first international conference on Networks for grid applications
Measuring TeraGrid: workload characterization for a high-performance computing federation

International Journal of High Performance Computing Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

The inevitable move from a single large scale server to a distributed Grid environment is beginning to be realized across international Grid test-bed like Pacific Rim Applications and Grid Middleware Assembly (PRAGMA). Although jobs submitted to a single large server have been widely analyzed, job characteristics in a Grid environment are different as we found in our analysis of jobs submitted in the PRAGMA Grid test-bed. This paper reports on the analysis of jobs submitted across the PRAGMA Grid test-bed. The job types are categorized and the runtime of jobs is captured, using the Multi- Organization Grid Accounting System (MOGAS), and analyzed. The number of jobs submitted across organizations, indicating the level of resource sharing among participants, is also captured by the system.