2013 Fiscal Year Final Research Report
Power-Saving Virtual Machine Scheduling for Many-Core CPUs in Cloud Environments
| Project/Area Number |
23500050
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Software
|
|---|
| Research Institution | Keio University |
|---|
Principal Investigator |
KONO Kenji 慶應義塾大学, 理工学部, 准教授 (90301118)
|
|---|
| Project Period (FY) |
2011 – 2013
|
| Keywords | クラウド環境 / 省電力 / 仮想化 / マイグレーション |
|---|
| Research Abstract |
The cost of electric power is becoming dominant in modern cloud environments. In the modern cloud environments, it is norm to use virtual machines to manage their physical resources. In this research, several techniques for saving electricity in data centers have been established. They include the technique to promote the virtual machine consolidation for power saving by means of live migration. It also reduces the power consumption by network switches. Another technique that allows us to select cost-effective data centers in terms of electricity prices. In addition to those techniques, the virtualization of GPU, an example of many-core processors has been achieved. It opens a way to power saving of many-core processors.
|
