How to allocate resources to 2 VMs on a dedicated server for maximum performance? [closed]

Question

Closed. This question is opinion-based. It is not currently accepting answers.

---

Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.

Closed 8 years ago.

Improve this question

I have a dedicated server with 32gb of RAM.

This server exists solely to host two virtual machines. (VMWARE vSphere) It does nothing else.

How much memory can I allocate to each of these machines for maximum performance?

Answer 1

Not sure I understand your question fully, but I guess you mean: When I have 32GB RAM on my ESXi host, how much is left to run VMs in?

For running ESXi on a 32GB host, my guess is that you will lose about 500MB to the hypervisor.

Per VM there will be some memory overhead, which depends on number of vCPUs and RAM assigned, see: http://pubs.vmware.com/vsphere-4-esx-vcenter/topic/com.vmware.vsphere.resourcemanagement.doc_41/managing_memory_resources/r_overhead_memory_on_virtual_machines.html

When the VMs are NOT 64bit and are NOT using Large Pages, you will have Transparent Page Sharing which means memory dedupe. If duplicate small pages of memory (4K) are found in physical ESXi memory, the page is only stored once. This can save quite a lot of memory.

Read my blogpost on memory overcommit: http://www.gabesvirtualworld.com/memory-overcommit-in-production-yes-yes-yes/ Also read on memory compression in ESXi. Search my blog for "Memory management and compression" (I can only post two links in this answer because I don't have enough reputation yet.

If you read the links, you'll understand why there is not just one answer :-)