I feel that Turbonomic is moving VMs around far more than what is necessary given our fairly low resource utilization levels. Is there a report that drills down into why Turbonomic is making its VM move decisions? Is it because of host network traffic? Storage I/O? CPU? Clearly Turbonomic "sees" benefit in moving these VMs to different hosts but I would like some help in identifying the root causes.
Our vMotion runs on a physically separate 1Gb network that consists of multiple switches. Several of these switches have only a single 1Gb switch-to-switch link! We're moving to correct this but it'll be a few months before anything happens. In the interim, vMotions are somewhat expensive operations and I'm trying to limit them to situations where a clear and discernible benefit would be realized -- not just better looking numbers on a screen!