A. In my last question here, I talked about the use of memory priority in Hyper-V R2 SP1's dynamic memory feature. Memory priority is an important feature that protects high-priority virtual machines (VMs) when physical memory runs out on the host. But in that article, I discussed that any time memory priority rules need to be referenced, you're in a situation where there's not enough memory to go around.

If you look at the list of VMs in Hyper-V Manager, you'll see that SP1 adds two new columns to the list: Current Memory and Memory Available. On a healthy system where there's enough RAM to be distributed to needy VMs, the counter for Memory Available should show a small percentage value.

This value represents the percentage of RAM that's currently available to the VMs, based on the total RAM that's been assigned at that moment. Both of these numbers will constantly change as VM memory requirements change. In a healthy system—one with enough RAM to go around—this value should be fairly close to the amount of buffer you've configured for the VMs (see the answer to this question for a full description of how this works).

You can create a situation where this number becomes a negative value. When the Memory Available counter is a negative value, this means that the sum total of memory needs by all VMs on the host is greater than the physical RAM available on that system. In such a situation, a Hyper-V host's memory balancer will remove memory from VMs that need it for their assigned jobs.

This removal of memory follows the memory priority rules (which were discussed here). The negative value tells you how much memory that VM needs but isn't receiving because the host is overcommitted.

Catch up with @ConcentratdGreg on Twitter!