Reactive Fault Tolerance aware Workflow Scheduling Technique for Cloud Computing using Teaching Learning Optimization Algorithm

Abstract

Cloud computing has led to a metamorphosis within the delivery model of information technology, from a product to a service. The resources of cloud computing should operate without mistakes or malfunctions is one of the primary challenges. Nevertheless, the cloud environment's ever-changing characteristics lead to a variety of unforeseen problems and malfunctions. Cloud services are hampered by their own vulnerability to disruption as a result of faults or failures in operations, causing them to perform less well than they could. To ensure the availability and dependability of crucial resources, fault tolerance is a fundamental concern. Failures should be anticipated and handled proactively. In order to achieve resilience and reliability in cloud computing, effective failure evaluation and management are required. To anticipate these problems and take the right action before they really happen, many proactive fault tolerance techniques have been proposed and are being deployed. However, most of the proactive fault tolerance methods could not yield a significant solution or help anticipate the faults. Failure to complete a task is no longer an accident but rather a typical feature in cloud computing environments. In this paper, a reactive fault tolerance-aware workflow scheduling approach (RFT A WS) for cloud environment is proposed, that is optimized for teaching and learning. This approach aims to reduce the likelihood of dynamic tasks failing before they are supposed to by taking into account the resources that are already available. The effectiveness of the suggested approach RFT A WS is assessed using the failure ratio, performance improvement rate, and rejection ratio to estimate the scheduling task. The result shows that the suggested system works better than the ones that are already in place and makes data more accessible and reliable.