MAS-Based Approach for Scheduling Intensive Workflows in Cloud Computing

• Published in: 2018 IEEE 27th International Conference on Enabling Technologies: Infrastructure for Collaborative Enterprises (WETICE) pp. 15 - 20
• Main Authors: Mokni, Marwa, Hajlaoui, Jalel Eddine, Brahmi, Zaki
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2018
• Subjects: Cloud computing; Multi agent system; QoS; Workflow scheduling
• DOI: 10.1109/WETICE.2018.00010

Cloud Computing is becoming popular model for delivering Information Technology (IT) that offers different services on demand over the Internet. This technology is dedicated to distribute computing resources and their consumption as software services. With the availability of data gathered from sophisticated scientific tools, workflows have proven their utility to implement relevant scientific realizations. Scheduling algorithms are primordial to efficiently automate these intensive workflows and many attemps have been made to develop new heuristics relying on a Cloud resource model. The majority of these heuristics address one or two-dimensional QoS issues. Instead, our work considers multi-dimensional QoS metrics for scheduling namely, execution time, cost, reliability, availability and ensures maximum balancing and exploitation of resources in order to minimize the energy consumption. In this paper, we propose a multi-agent approach to schedule intensive workflows in Cloud computing in such a way that all tasks will be executed with minimal possible time and smallest energy consumption. Our approach considers multi-dimensional QoS metrics for scheduling namely, execution time, cost, reliability, availability and ensures maximum balancing and exploitation of resources in order to minimize the energy consumption.