HDECO: A method for Decreasing energy and cost by using virtual machine migration by considering hybrid parameters

Energy and cost are important issues in cloud computing, which led to increased power supply consumption and more carbon dioxide production. In this paper, cost and energy have been studied with other methods and ECIB algorithm. By the study of the ECIB and previous methods, the HDECO method is pres...

Full description

Saved in:
Bibliographic Details
Published in:Computer communications 2022-11, Vol.195, p.49-60
Main Authors: Delavar, Arash Ghorbannia, Akraminejad, Reza, Mozafari, Sahar
Format: Article
Language:English
Subjects:
Energy
HDECO
ITD
Load-balancing
Virtual machine migration
Workflow scheduling
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Energy and cost are important issues in cloud computing, which led to increased power supply consumption and more carbon dioxide production. In this paper, cost and energy have been studied with other methods and ECIB algorithm. By the study of the ECIB and previous methods, the HDECO method is presented. This paper provides a method to decrease energy and cost in instance-intensive cloud workflows, using virtual machine migration. In the proposed method, creating a classification of inputs, calculating real execution time, and distance parameters, as well as using an intelligent threshold detector (ITD) compared to the study of the previous methods, has improved cost and energy, which according to the specified parameters, an optimum solution is achieved. In the HDECO method, energy has been improved by creating an objective function and combining parameters into the fitness function. By observing the provided parameters, the actual execution time is improved, and also dynamic threshold is used in the proposed method to reduce energy and cost. In this method, the accuracy of prediction increased by classification of inputs to different levels, as well as applying the ITD to the previous tasks and the ECIB algorithm. In our method, the allocation of the specified parameters such as ITD, the actual execution time, and the number of migrations have changed compared to the previous methods and ECIB. Finally, by categorizing the inputs related to the processing power of the resources as well as the proper allocation of them to a resource that has less work and suitable load-balancing, the execution time is optimized and the amount of energy consumption and cost is improved.
ISSN:0140-3664
1873-703X
DOI:10.1016/j.comcom.2022.08.006