Summary
Investigate if we can generate traces when the IOPS are getting slow. We need to define what "slow" is.

Motivation

• Does this affect any team outside of WT?
  It would be great for the support team to know when there are I/O issues. 
• How likely is it that this use case or problem will occur?
  Depending on the workload and the type of disk, those traces could help the investigation.
• Is this issue urgent?
  The earlier the better.

Acceptance Criteria (Definition of Done)

• Testing
  Make sure traces are generated.

• Documentation update
  Probably none