Rack Server Affinity Issues - Seeking Advice from the BadVoltage Community

Hi Everyone

I have a rack server setup with multiple nodes, each running different virtualized workloads. The idea is to optimize performance by assigning specific workloads to nodes that have the required resources. I’ve been using affinity settings to ensure that certain VMs run on specific nodes to make the best use of available CPU, memory, and other resources.

However, it seems like these affinity settings are not working as expected. Some virtual machines are not adhering to their assigned nodes and are migrating around, causing unexpected spikes in resource utilization and degraded performance. It’s almost as if the affinity rules are being ignored intermittently.

What I’ve Tried:

Double-checked my affinity settings: I revisited the configurations to make sure I haven’t overlooked any mistakes in assigning affinity rules to VMs.

Resource monitoring: I’ve closely monitored resource usage and migration patterns to identify any patterns that might indicate why the affinity rules are not holding up.

My Questions:

Has anyone else experienced similar issues with rack server affinity settings? How did you go about diagnosing and resolving the problem?

Are there any best practices or tips for configuring affinity settings on rack servers that I might be missing?

Could there be any underlying hardware or software issues that might cause affinity rules to be disregarded intermittently?

Thanks in advance for any advice, suggestions, or anecdotes you can share. Your support means a lot to me, and I’m looking forward to hearing your thoughts on this matter.

Best regards,

kinda difficult to help without knowing what virtualisation platform you’re using, what guest OSes, what CPUs you’re using (and are they the same on each virt host?), what shared storage you’re using, etc.