Research Area:  Cloud Computing
Cloud computing has become a well-known platform for solving big data and complex problems such as workflow applications. Infrastructure as a Service (IaaS) from the cloud is a suitable platform to solve these problems as it can potentially provide a nearly unlimited amount of resources using virtualization technology with a pay-per-use cost model. Various Quality of Service (QoS) objectives, such as cost and time, have been considered individually for workflow scheduling. In this paper, we proposed two energy-efficient heuristic algorithms with budget-deadline constraints that are appropriate for resources with Dynamic Voltage and Frequency Scaling (DVFS) enabled, as well as those that do not support DVFS. They are Budget Deadline Constrained Energy-aware (BDCE) and Budget Deadline DVFS-enabled energy-aware (BDD) algorithms for the cloud. Furthermore, they acquire affordable cost, faster scheduling length, and higher energy-saving ratio. Various evaluation metrics like success rate, cost and time ratios, energy consumption, utilization rate, and energy-saving ratio are utilized to evaluate the performance of the proposed algorithms. The obtained results are compared with budget-deadline constraints methods, such as BDSD, DBCS, and BDHEFT, as well as two other energy-efficient deadline-constrained algorithms, namely, ERES and Safaris algorithm in various scenarios on scientific workflow applications.
Author(s) Name:  Ahmad Taghinezhad-Niar, Saeid Pashazadeh & Javid Taheri
Journal name:  Computing
Publisher name:  Springer
Volume Information:  Volume 2022
Paper Link:   https://link.springer.com/article/10.1007/s00607-021-01030-9