Best posts made by tjkreidl

@olivierlambert said in NUMA-impact - Xeon/Epyc - 1P vs 2P:

There is no universal answer (because it's mostly depending on your VM load and what do you expect). As usual, my advice is to keep it simple if you don't have a problem with it (ie: you are satisfied by the perf.). Even a default EPYC configuration will be likely always better than a Xeon one.

After that, if you want to go deeper and learn the details, it's OK, let me just ping @tjkreidl who did a remarkable job (if I remember correctly) on this very topic.

Thanks for the mention, @olivierlambert ! Here's a link to part 3, which contains links back to parts 1 and 2. Note that NUMA will affect EPYC processors differently as they changed the die configuration at one point with the number of cores. I'm open for any questions on this topic. https://blogs.mycugc.org/2019/04/30/a-tale-of-two-servers-part-3-the-influence-of-numa-cpus-and-sockets-cores-persocket-plus-other-vm-settings-on-apps-and-gpu-performance/

@epretorious I would add that you have to be careful about overprovisioning when NUMA/vNUMA kicks in, that is when you allocate more VCPUs to exceed the number of physical CPUs of a bank of them as well as the associated physical memory (assume, for the sake of argument, you have two banks of physical CPUs and each has directly accessible to it one of two banks of memory) then things get inefficient because a CPU may need to go across to a different bank of memory to access data and there is additional overhead involved. See for example this article and the two preceding it:
https://blogs.mycugc.org/2019/04/30/a-tale-of-two-servers-part-3-the-influence-of-numa-cpus-and-sockets-cores-persocket-plus-other-vm-settings-on-apps-and-gpu-performance/

-=Tobias

@jhansen I'd suggest booting off some Linux rescue disk and just redoing all the partitions manually. I ran into an upgrade issue once on a Dell box because of some stupid little extra utility partition they had in there. See: http://xenserver.pl/author/tobias-kreidl
Look for "The (In)famous Dell Utility Partition."

Tapdisk is just palin not good as a storage control mechanism, IMO. There's only so much you can pull out of it.

@Andrew Thank you for that, much appreciated. Although I was aware of this process for SSD drives, I did not know that spinning disks had become that much smarter in the interim (~40 years!). But in any case, raw drives are very powerful if you have decent code to access them and the overhead can be appreciably less than with formatted drives.

@gustavoninetyone -- it's not on the supported hardware list, hence highly unlikely. See Olivier's response near the top of this category to see what GPUs are supported.
-=Tobias

@olivierlambert Indeed. NVIDIA in particular is bent on your having to license any sort of vGPU technology and it would not surprise me to see even passthrough leveraged at some point in the future. There are already restrictions on being able to make use of Quadro features, for example, of you don't have licenses.

Also, note that you need one video board reserved to run the native OS (XCP-ng) dom0 instance itself, so the server has to effectovely have at least two graphics boards such that one can be used as a GPU.

@Berrick Run iostat and see if anything shows up as slow or saturated with your storage I/O.

@gsrfan01 Not sure that was ever officially supported, though I did implement a direct connection to a VM on XenServer with better results than with the standard process. You're right, it may or may not work with much newer versions of XCP-NG/XS/CH.

@vsaad A PERC-6 is pretty old technology. Getting in the 50 MB/sec I/O range seems pretty good and some of the limitations will be on the disks, themselves. You might try running "iostat -x" while the Crystal Mark test is running to see if anything saturates, like the CPU, or if you get high wait states or other tell-tale signs,

@planedrop Once the VM is up and has XenTools installed, you should be able to add a lot of disks to it. I ran some VMs with at least 6 or 8 VDIs, as I recall. For fun, I think I added round 20 to a test VM!

@MichaelCropper CPUs can be over-provisoned, but not memory. You can use DMC (dynamic memory control) to regulate how much memory a VM will actually use, but in total, you still cannot exceed the total amount of physical memory available on a server.

CPU over-provisioning is very common, especially if loads change significantly over time (day/night weekday/weekend, special event and holidays/regular days, etc.).

Watching the load with top and xentop will give you an idea about overall performance of dom0 and all VMs, respectively.

As to a VM powred off, it will use up neither memory nor CPU resources.

There are a lot of subtleties involved that would entail a much longer discussion, but hopefully this will help for starters. You can google a lot of information about memory and VCPU allocation; there is a lot of information out there.

And from the CLI:

1. xe host-list (to get the UUID of the host)
2. xe pool-eject host-uuid=<host_UUID>