An agile framework adaptive to complicated memory workloads for VM migration
| Autor: | Yuqing Lan, Tao Han, Simin Yu, Chaoying Wu |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Service (systems architecture)
business.industry Computer science Distributed computing Improved algorithm 020206 networking & telecommunications Cloud computing Workload 02 engineering and technology computer.software_genre 01 natural sciences 010104 statistics & probability Prediction algorithms Virtual machine 0202 electrical engineering electronic engineering information engineering 0101 mathematics business computer Agile software development Live migration |
| Zdroj: | 2017 6th International Conference on Computer Science and Network Technology (ICCSNT). |
| DOI: | 10.1109/iccsnt.2017.8343474 |
| Popis: | A vital advantage of virtual machines (VMs) is live migration — the ability to transfer VMs from one physical machine to another as the VMs continue to offer service. Some well-known techniques have been proposed for live migration, such as pre-copy and post-copy. Unfortunately, these classical techniques are not agile enough in the face of complicated workloads since they are designed to work well in a specific workload. When the workloads of VMs get complicated, even a good migration algorithm may not run very well. This paper proposes an agile framework which determines the type of workloads of a VM by learning the statistics of usages of memory pages and then the framework chooses an appropriate improved algorithm to complete a VM live migration. The experimental results show that the framework is able to identify the type of workloads, and improves the performance of VM live migration. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|