On Selecting the Right Optimizations for Virtual Machine Migration

To reduce the migration time of a virtual machine and network traffic generated during migration, existing works have proposed a number of optimizations to pre-copy live migration. These optimizations are delta compression, page skip, deduplication, and data compression. The cost-benefit analysis of these optimizations may preclude the use of certain optimizations in specific scenarios. However, no study has compared the performance & cost of these optimizations, and identified the impact of application behaviour on performance gain. Hence, it is not clear for a given migration scenario and an application, what is the best optimization that one must employ? In this paper, we present a comprehensive empirical study using a large number of workloads to provide recommendations on selection of optimizations for pre-copy live migration. The empirical study reveals that page skip is an important optimization as it reduces network traffic by 20% with negligible additional CPU cost. Data compression yields impressive gains in reducing network traffic (37%) but at the cost of a significant increase in CPU consumption (5×). De-duplication needs to be applied with utmost care as the increase in CPU utilization might outweigh the benefits considerably. The combination of page skip and data compression works the best across workloads and results in a significant reduction in network traffic (40%).