Great guest post over at Virtualization.info by Kevin Lawton, Lead developer of Bochs on how to accelerate VMs live migration by 4-10x. In the post he points out that one of the keys to accelerating VM migration time is to make use of duplicate memory throughout the compute fabric. He also says that using relatively slow VM migration as a mechanism for scheduling, has a number of risks and short-comings, which leave its full potential untapped.
According to Lawton, "The time within which workloads can ebb and flow (and spike) is much quicker than the response time available to the scheduler to re-schedule VMs on other servers. As a result, the scheduler has to be ultra-conservative, otherwise it may break SLAs and/or create troublesome load-based hot-spots. By contrast, if the scheduler could expect near instantaneous VM migrations, it could perform much higher fidelity load-balancing or much more efficient power management (packing VMs onto absolutely the fewest number of powered-on servers). Thus, as live VM migration times decrease, the less conservatism is needed, and the greater the amount of potential performance and power savings can be wrung out of existing resources."
Very interesting and worth a read > http://www.virtualization.info/2009/04/r-accelerating-vms-live-migration-by-4.html