Hi
Can anyone explain what is the difference between the vsphere advance parameters “numa.autosize.vcpu.maxPerVirtualNode” and “numa.vcpu.maxPerVirtualNode”
Thanks
I would leave it, it seems to me that it overrides it anyway.
Thank you, you are right, I just tried it and below the output for before configuring ""numa.vcpu.maxPerVirtualNode" and after configuring. Just for your reference
Before:
numactl --hardware
available: 1 nodes (0)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11
node 0 size: 16891 MB
node 0 free: 7160 MB
node distances:
node 0
0: 10
After:
numactl --hardware
available: 2 nodes (0-1)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9
node 0 size: 14371 MB
node 0 free: 10072 MB
node 1 cpus: 10 11 12 13 14 15 16 17 18 19
node 1 size: 14613 MB
node 1 free: 9250 MB
node distances:
node 0 1
0: 10 20
1: 20 10
Frank has a deepdive post on his blog about NUMA and explains all the settings:
https://frankdenneman.nl/2016/08/22/numa-deep-dive-part-5-esxi-vmkernel-numa-constructs/
If I understand correctly then “numa.autosize.vcpu.maxPerVirtualNode” is configured based on the number of vCPUs a VM has. numa.vcpu.maxPerVirtualNode allows you to override it. At least that is how I interpreted the lengthy post.
Thank you very much, So if we add configuration parameter "“numa.vcpu.maxPerVirtualNode” we might remove the parameter “numa.autosize.vcpu.maxPerVirtualNode” am I right?
I would leave it, it seems to me that it overrides it anyway.
Thank you, you are right, I just tried it and below the output for before configuring ""numa.vcpu.maxPerVirtualNode" and after configuring. Just for your reference
Before:
numactl --hardware
available: 1 nodes (0)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11
node 0 size: 16891 MB
node 0 free: 7160 MB
node distances:
node 0
0: 10
After:
numactl --hardware
available: 2 nodes (0-1)
node 0 cpus: 0 1 2 3 4 5 6 7 8 9
node 0 size: 14371 MB
node 0 free: 10072 MB
node 1 cpus: 10 11 12 13 14 15 16 17 18 19
node 1 size: 14613 MB
node 1 free: 9250 MB
node distances:
node 0 1
0: 10 20
1: 20 10