A Study of Virtual Machineries Technology in Cloud Computing
| Main Author: | Omkar Ramesh Ghatage |
|---|---|
| Format: | Article Journal |
| Terbitan: |
, 2020
|
| Subjects: | |
| Online Access: |
https://zenodo.org/record/3784734 |
Daftar Isi:
- Cloud computing, along with social, mobile and analytic technologies, has revolutionized the Information Technology (IT) industry by enabling elastic on-demand provisioning of computing resources. According to Right Scale Survey, 82% of enterprises reported a cloud usage strategy as compared to 74% in 2014, which shows a 9.76% growth in cloud usage. On the other hand, predicts that there will be an increase of 41.3% of cloud usage at the end of 2016. The statistical report of indicates that more than 82% of the companies have envisaged tremendous cost benefits after migrating to cloud environment. The same report also estimates that that the global DC traffic will grow three-fold, while global traffic growth will increase by 3.5 fold in the next year. These statistics and predictions show cloud as an essential strategy with high service quality expectations in terms of safety, easy accessibility, cost and maximum resource utilization.In cloud computing, one area of research that has attracted researchers, industrialists and academicians equally is “Resource Management”. According to a data center uses four stages to perform efficient management of resources. They are, VM provisioning, resource provisioning (includes mapping and scheduling requests onto PMs), run-time management and workload modeling. In this research work, the focus in on VM provisioning, which apart from mapping VMs to appropriate PMs should alsotry to satisfy both client SLA for Qos and reduce operating costs, which is a challenging task for cloud service providers. More often, the client service providers are faced with challenges of under-provisioning (a starvation or saturation of VM resources that leads to service degradation) and over-provisioning (underutilization and subsequent waste of VM resources). Under-provisioning often leads to SLA penalty resulting into business revenue loss on the part of the cloud providers and also a poor Quality of Experience (QoE) for the cloud client’s customers (unacceptable response time for time critical applications for example). On the other hand, over-provisioning can lead to excessive energy consumption, culminating in high operating cost and waste of resources.