JetStress (WR2.1) uses Database Page Fault Stalls Per Sec instead of Database disk write latency.
It's mainly because it could handle SRDF/sync environment. In SRDF/sync environment, database disk write latency can go over 20 ms on average, but it is not necessarily causing users to experience performance problems. However, in a non-replicated environment, if the database write latency is over 20 ms, it is very likely the storage is having a performance problem.
The Exchange 2003 performance troubleshooting guide still has the 20 ms average database write latency threshold for a non-replicated environment, but will have zero database page fault stalls/sec threshold for a replicated environment.
In JetStress, we could not easily detect the environment, so that we could not help relying on database page fault stalls/sec for simplicity reasons.
This posting is provided "AS IS" with no warranties, and confers no rights.