A PSO task scheduling and IT2FCM fuzzy data placement strategy for scientific cloud workflows

Abstract

Scientific workflows are appropriate for treating large volumes of data and leading large-scale experiments. They are time-consuming and resource-intensive applications that profit from operating within distributed platforms, like cloud computing. However, the cloud presents several challenges that must be addressed to employ workflow applications efficiently. Although the workflow scheduling problem has been broadly examined, it still attracts attention. Furthermore, in the literature, little research has been conducted taking into consideration the data placement process in parallel with the execution of task scheduling. Hence, this paper proposes a task scheduling and data placement strategy to reduce the number of data movements between cloud data centers. The suggested strategy uses an algorithm based on the fuzzy clustering method Interval Type-2 Fuzzy C-Means (IT2FCM) and the meta-heuristic optimization technique Particle Swarm Optimization (PSO) to minimize data movements throughout the workflow’s execution. The proposed strategy is evaluated through a simulation process using various well-known scientific workflows of different sizes. Additionally, the outcomes show that our strategy is more reliable than the most recent state-of-the-art techniques.