Performance evaluation of Static Level based Batch Scheduling Strategy (SLBBS) for computational grid

Abstract

Computational grid is a parallel and distributed infrastructure involving heterogeneous resources providing dependable, pervasive and consistent access of compute intensive resources from multiple organizations that are not subject to centralization at the administrative level delivering optimized QoS parameters. Job scheduling (i.e. mapping) is the core issue in computational grid. The problem of mapping jobs onto heterogeneous computational resources to optimize one or more QoS parameters has been proven to be NP-Complete. This work discusses the Static Level based Batch Scheduling Strategy (SLBSS) and other state of art batch scheduling algorithms viz. Min Min, Max Min, Sufferage and LJFR-SJFR along with their design motivation and limitations. A performance evaluation and analysis of SLBSS with its peers is done to evaluate its significance in the middleware.