Reducing Resource Idle Time in Virtualized Cloud Data Centersby Taking Advantage of a Hybrid Fuzzy-Genetic Method

سال انتشار: 1403
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 63

فایل این مقاله در 13 صفحه با فرمت PDF قابل دریافت می باشد

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

ICECM09_057

تاریخ نمایه سازی: 27 اسفند 1403

چکیده مقاله:

cloud computing implies expanding and employing the technology of internet-network-basedsystems. This kind of computing includes the methods of system computing in the environment in whichthe abilities related to Information Technology (IT) are expressed as services for users and make themable to get access to the technology-based services in the internet network without having specializedinformation about these technologies or controlling the infrastructures of the technology supportingthem. There are different problems to use cloud computing environment—one of the most importantproblems is Virtual Machine (VM) placement in the data centers in this environment. The placement indata centers should be done more accurately due to the different conditions of the virtual machines in acloud computing environment including VM Memory Usage and VM CPU Performance. To deal withthis challenge, we use a fuzzy-genetic method to locate the virtual machines in data centers. Bydesigning a fuzzy-genetic algorithm to locate virtual machines in data centers, the virtual machines canbe located based on the cloud computing capacity for each cloud resource. The proposed method in thisstudy is analyzed based on the simulations in MATLAB software and designed datasets. The obtainedresults show that the proposed method reduces the response time and idle time of the resources that leadto the more accurate placement of virtual machine and increased service quality in a cloud computingenvironment. The extent of response time and idle time reduction in the proposed virtual machineplacement method is extracted and compared with the CPVMP and CRVMP virtual machine placementmethods in cloud data centers. The final results indicate the reduction of the mentioned metrics in theproposed method due to the more accurate virtual machine placement in cloud data centers.

نویسندگان

Abazar Barzegar

Department of Computer, Faculty of Sciences, Neyriz Branch, Islamic Azad University, Neyriz, Iran