Reliability-Based Heuristic-Based Algorithm for Task Execution in Multiprocessor Systems

Abstract
Objective: This study implements a combination of heuristicbased algorithm and finding variable neighborhoods, thereby reducing make-span and improving reliability.

Methods/statistical analysis: The bi-objective algorithm is proposed for a static planning strategy to achieve high performance in heterogeneous multiprocessor systems. The reliability of a system is based on the probability in which resources of the system execute tasks without any failure.

Findings: Here, a genetic algorithm integrated using single neighborhood structure Genetic Variable Neighbourhood Search (GVNS) is implemented to improve the efficient search quality.

Applications and improvements: Simulation is performed to maintain better performance parameters when compared with conventional algorithms.

Keywords: Directed Acyclic Graph, Inter Process Communication, Evolutionary Algorithm.