Proceedings of the 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017)

Live Virtual Machine Migration with Bandwidth Dynamic Assignment

Authors
Yiqiu Fang, Fangzhengqing Liao, Junwei Ge
Corresponding Author
Yiqiu Fang
Available Online April 2017.
DOI
https://doi.org/10.2991/fmsmt-17.2017.161How to use a DOI?
Keywords
virtual machine, dynamic migration, bandwidth dynamic allocation, dirty page rate, cloud computing.
Abstract
In this paper, we present an algorithm that based on the probability prediction of the dirty page's Pre-Copy algorithm, bring in a prediction algorithm of the bandwidth required for the transmission of the dirty pages. Mainly in the case of mutations of dirty page, Analysis of the probability of occurrence of the dirty page in the process of migration, Obtain the adjusted bandwidth coefficient, dynamically allocates the bandwidth required for the current transmission, and reserved for later transmission bandwidth, so that the transmission process will not have the use of bandwidth imbalance or the bandwidth cannot make full use of, so that each time from the current virtual machine to reach the directory virtual machine transmission speed can be consistent as much as possible. Experiments show that the algorithm can reasonably use limited bandwidth resources, reduce iteration time and downtime, and improve the dynamic migration performance of virtual machines in the case of mutations.
Open Access
This is an open access article distributed under the CC BY-NC license.

Download article (PDF)

Cite this article

TY  - CONF
AU  - Yiqiu Fang
AU  - Fangzhengqing Liao
AU  - Junwei Ge
PY  - 2017/04
DA  - 2017/04
TI  - Live Virtual Machine Migration with Bandwidth Dynamic Assignment
BT  - Proceedings of the 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017)
PB  - Atlantis Press
SP  - 835
EP  - 841
SN  - 2352-5401
UR  - https://doi.org/10.2991/fmsmt-17.2017.161
DO  - https://doi.org/10.2991/fmsmt-17.2017.161
ID  - Fang2017/04
ER  -