Fuzzy Logic Based Re-Engineered Model for Maximization of Average Processor Utilization of Distributed Real-Time System

Abstract

High performance parallel applications are using distributed real time system [DRTS] as phenomenal platform. DRTS process the parallel applications over multiple processors of the system. Performance of the system can be amended by efficient and evenly allocation methodology for parallel application’s tasks over the processors available in the system. Task allocation is NP-hard or NP-complete problem. Fuzzy logic based and dynamic load sharing technique based model for the distribution of parallel application’s tasks has been proposed and deployed in this paper. In the new heuristic suggested in model, the limitation of memory has been deployed using fuzzy rules and the triangular member function with its FIS variables has been used. In the proposed model k-mean clustering is being used and MATLAB 7.11.0 has been used to simulate the proposed model.