MECOM: Live migration of virtual machines by adaptively compressing memory pages

• Published in: Future generation computer systems 2014-09, Vol.38, p.23-35
• Main Authors: Jin, Hai, Deng, Li, Wu, Song, Shi, Xuanhua, Chen, Hanhua, Pan, Xiaodong
• Format: Article
• Language: English
• Subjects: Compression algorithm; Live migration; Pre-copy; Virtual machine
• ISSN: 0167-739X; 1872-7115
• DOI: 10.1016/j.future.2013.09.031

Live migration of virtual machines has been a powerful tool to facilitate system maintenance, load balancing, fault tolerance, and power-saving, especially in clusters or data centers. Although pre-copy is extensively used to migrate memory data of virtual machines, it cannot provide quick migration with low network overhead but leads to large performance degradation of virtual machine services due to the great amount of transferred data during migration. To solve the problem, this paper presents the design and implementation of a novel memory-compression-based VM migration approach (MECOM for short) that uses memory compression to provide fast, stable virtual machine migration, while guaranteeing the virtual machine services to be slightly affected. Based on memory page characteristics, we design an adaptive zero-aware compression algorithm for balancing the performance and the cost of virtual machine migration. Using the proposed scheme pages are rapidly compressed in batches on the source and exactly recovered on the target. Experimental results demonstrate that compared with Xen, our system can significantly reduce downtime, total migration time, and total transferred data by 27.1%, 32%, and 68.8% respectively. •To the best of our knowledge, we are the first to exploit memory compression technique to minimize the downtime and total migration time of live migration.•By analyzing memory page characteristics, we implement an adaptive memory compression approach to tune migration performance of virtual machines. A zero-aware characteristics-based compression (CBC) algorithm is designed for live VM migration.•Memory compression is adaptive to network bandwidth, then the performance of live VM migration still keeps good in diverse network environments with different available network bandwidth.•Experiment results demonstrate that compared with Xen, our approach can significantly reduce downtime, total migration time, and total transferred data by 27.1%, 32% and 68.8% respectively.