Statistical-Based Heuristic for Scheduling of Independent Tasks in Cloud Computing

Abstract

Cloud computing is an emerging and innovative technology that is used for solving large-scale complex problems. It considers as an extension to distributed and parallel computing. Additionally, it enables sharing, organizing and aggregation of computational machines to satisfy the user demands. One of the main goals of the task scheduling is to minimize the makespan (i.e. the overall processing time) and maximize the machine utilization. This paper addresses the problem of how to schedule many independent tasks when using different machines. It introduces two batch mode heuristics algorithms for scheduling independent task in the computational cloud environment, high mean absolute deviation first heuristic and QoS Guided Sufferage-HMADF heuristic. Besides, the paper presented other existing batch mode heuristics such as, Min-Min, Max-Min and Sufferage. The four heuristic modes are simulated and the experimental results are discussed using two performance measures, makespan and machine resource utilization.