Volume 18, No. 6, 2021
Efficient Fault Tolerant Cost Driven Mechanism For Scientific Workflow Through Optimal Replication Strategy In Cloud Computing Environment
Asma Anjum , Dr. Asma Parveen
Abstract
Cloud Computing has been one of the distributed and effective computing paradigms; it provides enormous opportunities to tackle the scientific problems that possesses large scale attribute. Despite of being such a flexible computing paradigm, it possesses several challenges and fails to achieve the required QoS. Reliability requirement is one of the most important quality of services (QoS) and should be satisfied for a reliable workflow in cloud computing. Primary-backup replication is an important software fault-tolerant technique used to satisfy reliability requirement. Recent works studied quantitative fault-tolerant scheduling to reduce execution cost by minimizing the number of replicas while satisfying the reliability requirement of a workflow on heterogeneous infrastructure as a service (IaaS) cloud. However, a minimum number of replicas does not necessarily lead to the minimum execution cost and shortest schedule length in a heterogeneous IaaS cloud. In this research work, we develop a ODS (Optimal duplication strategy) for fault tolerance and cost driven mechanism also named as ODS-FTC; ODS-FTC uses the iterative based approach that selects VM and its duplicates that has minimum makes pan in case of individual task. Moreover, this provides the utility against failure occurrence and optimal selection with optimal redundancy causes the cost to be optimal. ODS-FTC is evaluated considering the scientific workflow like cyber shake, LIGO, montage and SIPHT; evaluation is carried out through designing instances. Furthermore, in case of all instances, ODS-FTC is proved to be marginally improvised than the existing model.
Pages: 1955-1975
Keywords: Fault Tolerance, Reliability Requirement, Cost optimization, makes pan, efficient scheduling,