The fix appears to have worked, and performance is back to normal.

An offending service has been turned off, and we are monitoring the results.

We are investigating significantly higher latency in our database layer. This appears to be causing degraded performance in the web app for some users.