Task scheduling optimization based on heuristic algorithm for heterogeneous cloud computing platforms

Document Type : Research Paper


Department of Computer Engineering, Babol Branch, Islamic Azad University, Babol, Iran


In recent years, the issue of power consumption in parallel and distributed systems has attracted a great deal of attention. Regarding the ever-increasing  development data and computing centers due to the contribution of cloud computing systems in such sectors, power consumption has always been of the  concerns due to Carbon dioxide emissions and consequently the Negative impact on the environment. In recent years, the notion of power and also "Green  Computing" has found a crucial spot in the tasks scheduling in cloud data centers. The clustering technique, as well as Dynamic Voltage and Frequency  Scaling (DVFS) techniques, have focused on the reduction of the consumption of power particularly, and the optimization of the performance parameters.  Concerning scheduling Directed Acyclic Graph (DAG) of a data center processors equipped with the technique of DVFS, this paper proposes a power and time  aware algorithm called PATCDD, to apply the combination of the strategies for clustering along with the distribution of slack-time among the tasks of a  cluster. The first phase studies the slack time for non-critical tasks of DAG, extends their execution time and reduces the energy consumption without increasing the task’s execution time as a whole. The main idea of the proposed algorithm involves the achievement of a maximum reduction in power  consumption in the second phase. To this end, the slack time is distributed among non-critical dependent tasks. Eventually, a set of data established for  conducting the examinations and also different parameters of the constructed random DAG were assessed to identify the efficiency of our proposed  algorithm.    


Volume 13, Issue 2
July 2022
Pages 2743-2750
  • Receive Date: 02 September 2021
  • Revise Date: 16 March 2022
  • Accept Date: 20 April 2022
  • First Publish Date: 06 May 2022