1. مد ی لر ی ت آگاه به انرژی جهت زمان بندی وظا ی ف بلا استفاده از تکن ی ک مق ی اس بند ی پو ی ا ی ولتاژ و فرکانس در مراکز داده ابر ی – مطالعه موردی سازمان بنادر و در ی انورد ی.
- Author
-
د قربانی, م برزگر, and ی نظری
- Subjects
SERVER farms (Computer network management) ,PRODUCTION scheduling ,ENERGY consumption ,CLOUD computing ,RESOURCE management - Abstract
The issue of energy management is considered as one of the main concerns in cloud computing systems to support the rapid growth of data centers and computing centers Becomes. Dynamic voltage and frequency scaling strategies focus on reducing energy consumption as well as optimizing efficiency parameters. In this paper, we present an energy-conscious and reliable three-step scheduling algorithm called ERADVFS using the DVFS technique on a cloud center processor with dynamic voltage and frequency scaling capability. The goal of this algorithm is to consume the least amount of energy while ensuring reliability and limited task scheduling. In the first step, we determine the worst execution time of each task in each virtual machine of that type on each frequency level. Then in the next step, we divide the deadline for doing the work between the tasks. Finally, by assigning the DVFS technique, we assign tasks to the best virtual machine. The main idea in selecting a virtual machine intelligently by the DVFS technique is to take advantage of the extended amount of runtime to reduce the frequency of the virtual machine. Finally, to prove the effectiveness of the above algorithm in applications in the real environment, we have performed a series of experiments on a real production process of the Ports and Maritime Organization. The results showed that the above algorithm works better than other methods while maintaining the reliability and time required. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF