| 한글제목(Korean Title) |
Predictive Consolidation of Virtualized Resources Towards Performance and Energy Efficient Cloud Data Centers |
| 영문제목(English Title) |
Predictive Consolidation of Virtualized Resources Towards Performance and Energy Efficient Cloud Data Centers |
| 저자(Author) |
Frank I.
Elijorde
Jaewan Lee
|
| 원문수록처(Citation) |
VOL 14 NO. 02 PP. 0057 ~ 0058 (2013. 11) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
The dynamic usage pattern characterized by the pay-per-use model of cloud computing has made it clear to service providers the necessity of consolidating virtualized resources within the agreed upon SLA. Originally, the primary concern regarding the creation and provisioning of on-demand services was limited to the physical infrastructures of cloud data centers. However, as computing hardware have dramatically improved, an emerging dilemma among cloud data center operators is to keep energy consumption at a minimum. Therefore, cloud service providers should consider optimization when allocating virtualized resources in order to balance system performance and energy consumption. We propose a predictive SLA-aware approach for consolidating virtualized resources in a cloud environment. In order to maintain the quality of service as well as to establish an optimal trade-off between performance and energy efficiency, the server’s utilization threshold can dynamically adapt to the physical machine’s resource consumption using predicted values generated by an ANN. We implemented our work through simulations and evaluation results show that our approach outperforms non-power aware methods that don’t support migration as well as those based on static threshold.
|
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
