Authors: Dang Minh Quan
Addresses: Paderborn Center for Parallel Computing (PC2), University of Paderborn, Fuerstenalle 11, Paderborn, 33102, Germany
Abstract: Service Level Agreements (SLAs) serve as a foundation for a reliable and predictable job execution at remote grid sites. In this paper, we describe an error recovery mechanism for workflow within the SLA context, coping with catastrophic failure when one or several High Performance Computing Centers (HPCCs) are detached from the grid system. We propose an algorithm to detect all affected sub-jobs when the error happens and an algorithm to remap those sub-jobs to the remaining healthy HPCCs with makespan optimise. The experiment result shows that our mechanism discovers a higher quality solution in a shorter time period than other existing methods.
Keywords: grid computing; service level agreements; SLA; error recovery; workflow; mapping; high performance computing.
International Journal of High Performance Computing and Networking, 2007 Vol.5 No.1/2, pp.110 - 121
Published online: 14 Nov 2007 *Full-text access for editors Access for subscribers Purchase this article Comment on this article