Reliable budget aware workflow scheduling strategy on multi-cloud environment

• Published in: Cluster computing 2022-04, Vol.25 (2), p.1189-1205
• Main Authors: Chakravarthi, K. Kalyana, Neelakantan, P., Shyamala, L., Vaidehi, V.
• Format: Article
• Language: English
• Subjects: Algorithms; Back up systems; Budgets; Cloud computing; Communication; Computer Communication Networks; Computer Science; Failure; Fault tolerance; Heuristic; Operating Systems; Processor Architectures; Provisioning; Resource allocation; Run time (computers); Scheduling; System reliability; Workflow software
• ISSN: 1386-7857; 1573-7543
• DOI: 10.1007/s10586-021-03464-4

The resource provisioning and workflow execution in a multi-cloud environment using a pay-as-you-use framework have recently gained the attention of the cloud computing research community. Scheduling of workflows in the multi-cloud platform is challenging due to the cloud dynamics, particularly, heterogeneous resource types, multiple billing mechanisms, elasticity, on-demand provisioning, and systems reliability. In addition, these workflow applications have a runtime constraint—the most typical being the execution time and the execution cost. Another vital Quality of Service (QoS) metric that is of critical concern is reliability. This paper proposes a Normalization based Reliable Budget constraint Workflow Scheduling (NRBWS) algorithm to improve the workflow execution reliability and reduce the makespan under the budget constraint specified by the user. This scheme undergoes a min–max normalization process that is trailed by the computation of the expect reasonable budget ( erb ) to assign the tasks to one of the computational resources. The NRBWS algorithm lowers the makespan by assigning each workflow task to the most reliable computing resource with the earliest finish time under the allocated budget. Simulation results demonstrate that the proposed NRBWS algorithm outperforms existing state-of-the-art heuristics.