vMotion pre-copy rate

As described in VMware vSphere vMotion Architecture, Performance and Best Practices in VMware vSphere 5 Performance Study:

Because the virtual machine continues to run and actively modify its memory state on the source host during this phase, the memory contents of the virtual machine are copied from the source vSphere host to the destination vSphere host in an iterative process. The first iteration copies all of the memory. Subsequent iterations copy only the memory pages that were modified during the previous iteration. The number of precopy iterations and the number of memory pages copied during each iteration depend on how actively the memory is changed on the source vSphere host, due to the guest’s ongoing operations.

The number of precopy iterations also depend on the speed of the link(s) between the two hosts. Monitoring the pre-copy rate let you know if migration runs at full speed or not. The vMotionRate dashboard helps visualize the ratio of good/average/bad reported network bandwidth before the switchover. It is calibrated for 10GbE but you can easily change the values to fit your environement.


2015-02-20T11:05:06.449Z esx.vmware.com vmkernel: cpu3:427570)VMotion: 3878: 1424430291970314 S: Stopping pre-copy: only 3491 pages left to send, which can be sent within the switchover time goal of 0.500 seconds (network bandwidth ~931.354 MB/s, 1049920% t2d)




 This dashboard is embedded in SexiLog. If you want up-to-date version for your own platform, the Gist code for this dashboard is:  https://gist.github.com/sexibytes/6fac0bc3e4a1dabc377b
More information about dashboard import in Kibana is available here