Vladik Romanovsky

@xpivarc could you please add some words to the description? Just to understand how is this beating achieved and what are the tradeoffs.

/approve Thanks @xpivarc, looks good to me. I would really change title of this PR.

hmm... looks like the "local" provider was removed. https://github.com/kubevirt/kubevirtci/pull/1102 It did pretty much what you are asking for. I used it extensively in the very distant past.

@iholder101 maybe you want to take a look at the removed code and revive that?

@orelmisan is that part of VEP29? if so can you please add `VEP tracking issue: [link]` to the description?

@orelmisan Thanks! Looks good from my side as well. /approve

/test all

Thanks @lyarwood. Indeed, this is open to interpretation. So far, the general intent was to provide dedicated CPUs to the guest however, the topology assignment was on a best-efforts basis....

> ### node-labeller > node-labeller mainly reflects node specifications that do not change frequently, if ever, like supported CPU models, machine types, CPU features, etc. > > I think we...

> [@vladikr](https://github.com/vladikr) What do you mean by detection? We already rely, in case of node-labeller, on the init container that will just once. Do you suggest to change this? +1...