Hi folks.
I've been trying to understand this for some days but it's still not clear to me. Read some articles about hyper-v memory handling but didn't help much on my case.
In a hyper-v Cluster with 4 nodes (Windows Server 2016 Datacenter) I have a VM (let's call it VM1, also with WS 2016 DC) that has 128GB static assigned memory. Note: It was created as Gen1 VM and the memory assigned is not exactly 128GB, but 128000MB (don't ask me why it was created that way, but i thought maybe the info could be relevant)
VM1's Memory Demand (in Hyper-V Manager) is about 111GB (86%) but the guest OS is using less than 50% (RAMMap below, from Guest OS)
Got an error updating the image here. See it here: https://ibb.co/WsH1JD6
I'm not worried about the host memory, since the majority of the VMs are static assigned. My question is because I was using the perf counter "Memory Pressure" to monitor my VM memory usage and see if any of them are close to start paging. But with this case (Memory Pressure is around 86) I'm starting to think that counter might not be the best way.
First and main question: Why memory demand and Guest OS memory usage are so different?
Second: Does it make sense to use Memory Pressure to monitor a VM with static assigned memory?
If I'm misunderstanding any concept here, it would be nice to know.
Thanks!