Adaptive Fault Tolerant Task Scheduling Algorithm using Metaheuristics in Cloud Environments

Abstract

With pervasive network access, cloud computing offers on-demand access to a shared pool of computing resources. To guarantee high quality cloud services, cloud task scheduling algorithms need to be dependable and efficient. Cloud infrastructure, however, are vulnerable to errors and malfunctions that may affect how tasks are carried out. In this research, a metaheuristic approach for adaptive fault tolerant job scheduling in cloud systems is proposed.  The method efficiently schedules work while optimizing fault tolerance by adaptively applying numerous fault tolerance strategies and meta-heuristic based optimization. Additionally, a thorough analysis of modern Meta-heuristic based job scheduling algorithms is presented in this work, with an emphasis on the algorithm’s fault tolerance and adaptability. Through simulation tests using CloudSim toolkit, adaptability, scalability, energy efficiency, and reliability matrices are used to compare the suggested algorithm against existing approaches. The suggested methodology works better than current techniques, according to the results, which indicate increased task throughput, scalability, energy efficiency, and fault tolerance across a range of cloud workloads.  An efficient method for ensuring consistent task execution in unstable cloud environments is offered by the innovative scheduling architecture that incorporates six metaheuristics, redundancy, checkpointing and migration based fault tolerance in an adaptable manner.