Reliability driven resource Scheduling algorithms for grid Computing

Abstract

Grid computing has emerged as a new technology for solving newlinecomputational and data intensive problems within the scientific research newlinecommunity and industries such as medical weather forecasting space newlineexploration and the next generation e Science applications which require newlinelarge scale resource sharing wide area communication and multiinstitutional newlinecollaboration It will be critically important to develop methods newlinefor ensuring the reliability of large scale grid systems and to ensure that the newlinecreated specifications support the reliability of the system It is hard to newlineanalyze and model the Grid reliability because of its size complexity and newlinerigidity newlineThe motivation for this thesis work stems from the extensive newlineliterature survey made in scheduling and utilization of grid resources From newlinethe literature it has been observed that the probability of faults in grid newlineenvironment is much higher than the traditional distributed systems It is also newlinefound that very few researchers have analyzed the grid system reliability newlineMost of the GA based solutions presented in the literature do not consider the newlineQoS constraints such as redundancy budget and deadline Hence this thesis newlinework focuses on the direction of resource sharing and workflow scheduling newlineby taking into consideration the prior knowledge of computational resources newline newline