Your organization is baselining a Windows Server 2022 file server so that the monitoring system will alert any time disk latency stays above 25 ms for more than one minute. Which built-in Performance Monitor counter most directly reports the time the storage subsystem takes to complete each read or write request, allowing you to create a single-value latency threshold?
The Avg. Disk sec/Transfer counter records the average time, in seconds, that the disk subsystem needs to finish each I/O (read or write). Because it already represents latency as an elapsed time per transfer, you can compare the returned value directly to a millisecond threshold such as 0.025 s (25 ms).
Disk Bytes/sec and Disk Transfers/sec measure throughput and IOPS, not latency; high values may exist on both healthy and unhealthy arrays.
Current Disk Queue Length shows how many I/O operations are waiting, but interpreting it requires knowing how many spindles or virtual channels back the volume, so it is less precise as a stand-alone latency alarm.
Ask Bash
Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.
What is disk latency in servers?
Open an interactive chat with Bash
How does the Avg. Disk sec/Transfer counter measure latency?
Open an interactive chat with Bash
What is the difference between throughput and latency?