qm: multiqueue specify that it needs to be vCPUs not cores

total number of cores != vCPUs if there is more than one socket
configured. according to the redhat docs it should be vCPUs not cores:

> Multi-queue virtio-net provides the greatest performance benefit when:
> [..]
> - The number of queues is equal to the number of vCPUs.

https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/7/html/virtualization_tuning_and_optimization_guide/sect-virtualization_tuning_optimization_guide-networking-techniques#sect-Virtualization_Tuning_Optimization_Guide-Networking-Multi-queue_virtio-net
Signed-off-by: Stefan Sterz <s.sterz@proxmox.com>
This commit is contained in:
Stefan Sterz 2024-03-21 16:50:43 +01:00 committed by Aaron Lauterer
parent aefdf31cbc
commit 3d56535921

11
qm.adoc
View File

@ -770,14 +770,15 @@ vhost driver. With this option activated, it is possible to pass _multiple_
network queues to the host kernel for each NIC.
//https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/7/html/Virtualization_Tuning_and_Optimization_Guide/sect-Virtualization_Tuning_Optimization_Guide-Networking-Techniques.html#sect-Virtualization_Tuning_Optimization_Guide-Networking-Multi-queue_virtio-net
When using Multiqueue, it is recommended to set it to a value equal
to the number of Total Cores of your guest. You also need to set in
the VM the number of multi-purpose channels on each VirtIO NIC with the ethtool
command:
When using Multiqueue, it is recommended to set it to a value equal to the
number of vCPUs of your guest. Remember that the number of vCPUs is the number
of sockets times the number of cores configured for the VM. You also need to set
the number of multi-purpose channels on each VirtIO NIC in the VM with this
ethtool command:
`ethtool -L ens1 combined X`
where X is the number of the number of vcpus of the VM.
where X is the number of the number of vCPUs of the VM.
You should note that setting the Multiqueue parameter to a value greater
than one will increase the CPU load on the host and guest systems as the