Deadline based task scheduling using multi-criteria decision-making in cloud environment

• Main Authors: Suvendu Chandan Nayak, Chitaranjan Tripathy
• Format: Article
• Language: English
• Published: Elsevier 2018-12-01
• Series: Ain Shams Engineering Journal
• Online Access: http://www.sciencedirect.com/science/article/pii/S2090447917301417

Task scheduling within the deadline is more challenging due to different task parameters: start time, execution time, deadline and number of virtual machines (VMs). The OpenNebula platform supports Backfilling algorithm to schedule deadline based tasks. If there are similar types of tasks, then the scheduling performance should be enhanced. In this work, we propose, implementation of Multi-criteria Decision Making (MCDM) using VIKOR method to the existing backfilling algorithm. The VIKOR method is implemented as a decision maker to resolve conflicts among similar tasks. The mechanism finds the best ideal task among the similar tasks and ranks them for scheduling. Therefore, a number of tasks are scheduled. The performance is evaluated by analyzing the number of tasks scheduled, resource utilization and task rejection, which is quite better in comparison to the existing algorithm. Keywords: Deadline sensitive task, Scheduling, Backfilling, MCDA/MCDM, VIKOR