Here's my scenario -
Host A and Host B are both connected to Datastore 1 and Datastore 2 (separate arrays, but i don't think it matters in this case).
Both hosts are part of a DRS/HA cluster. Datastore 1 and Datastore 2 are in a SDRS cluster with Storage I/O control enabled.
Host A is showing "normal" (for my environment) latency on all datastores.
Host B is showing huge spikes in write latency on Datastore 2 and I can't figure out why. There are no VM's on Host B that use Datastore 2. This seems puzzling to me. Why the huge latency spikes when the host has no VM's accessing Datastore 2??
Also, Host A shows acceptable latency on Datastore 2. No unexpected behavior there.
Thoughts?