This is standard practice in some domains. Reserve 1 or 2 cores for admin and possibly for the interrupt daemon, and isolate/affinitize app processes to the remaining cores -- ideally giving more resources to your "hot" threads, and better resources (the core "closest" to your NIC) to your network thread, etc.

I'm not sure if this is a real issue; Normally (on gke for example) you can't use 100% cpu because of this.

If they saw the issue, then either they have not configured their nodes right, missconfigured them or perhaps run something very old?

I'm quite curious to see a proper test bench