In the last few days, we started noticing a performance issue on our BI sql server instance. The first thing that jumped out at me was our disk wait times. I use sp_whoisactive to get quick server overviews all the time, and it's not unusual to see PAGEIOLATCH_SH waits around 20-70ms listed. However, I was suddenly seeing waits of 200-1000ms, which I'd never seen before. I did some more digging around, but nothing seemed amiss. I talked with our SAN administrator, and after some quick profiling, it seemed that the disks were serving up way less information than they were/are capable of. This is still in the process of being resolved, but reads on the server feel markedly slow now across the board - even just doing little things like showing all tables for a database in SSMS is noticeably slower.
I've noticed one other thing, which is...