Availability of a system with permissible repair time : Hirofumi Kobayashi, Masafumi Sasaki and Shigeru Yanagi. Proc. 15th Symp. Reliab. Maintainab. (Union of Japanese Scientists and Engineers, Tokyo), 1 (1985)

Abstract

Grid computing aims to enable “resource sharing and coordinated problem solving in dynamic, multi-institutional virtual organizations (VOs)”. However, due to the nature of heterogeneous and dynamic resources, dynamic failures in the distributed grid environment usually occur more than in traditional computation platforms, which cause failed VO formations. In this paper, we develop a novel self-adaptive mechanism to dynamic failures during VO formations. Such a self-adaptive scheme allows an individual and member of VOs to automatically find other available or replaceable one once a failure happens and therefore makes systems automatically recover from dynamic failures. We define dynamic failure situations of a system by using two standard indicators: mean time between failures (MTBF) and mean time to recover (MTTR). We model both MTBF and MTTR as Poisson distributions. We investigate and analyze the efficiency of the proposed self-adaptation mechanism to dynamic failures by comparing the success probability of VO formations before and after adopting it in three different cases: (1) different failure situations; (2) different organizational structures and scales; (3) different task complexities. The experimental results show that the proposed scheme can automatically adapt to dynamic failures and effectively improve the dynamic VO formation performance in the event of node failures, which provide a valuable addition to the field.