Forum Discussion
Monitor duration doesn't match refresh ID
Hello all!
We are migrating to powerbi service with F64 capacity.
I setup a semantic model with incremental refresh for the last 10 days of data that have been working fine for a couple of days, but I noticed that on the monitor, it takes around 7 min to complete, whereas if I click on the semantic model, the refresh ID shows a total time of ~2 min
So, two questions: How can I read the results of the monitor? and why the refresh ID shows less time to update?
I'm pretty sure I'm reading it wrong, but how can I interpret the results shown?
Thanks in advance!
Hello !
Thank you for posting on Microsoft Fabric community.
The Monitor duration is the end-to-end wall-clock time (it includes waiting in the capacity queue, model warm-up...). The Refresh ID time shows only the actual processing once a slot was obtained (Power Query/partition processing + commit, plus query cache).
So if Monitor says 7 to 8 minutes but Refresh ID shows 1 to 2 minutes, the extra minutes are mostly queue or warm-up time because the capacity was busy or just cold.
It is better to use Monitor as the total user wait and Refresh ID as the engine compute time.
- Monitor duration
- includes total job time—like queuing, cache refresh, and post-processing.
- Refresh ID duration shows only the actual data refresh time (partition processing).
- That’s why monitor shows ~7 min, but refresh ID shows ~2 min.
Use monitor for capacity insights, and refresh ID for performance tuning. Let me know if you want a quick visual tracker for both!
- Monitor duration
3 Replies
Hello !
Thank you for posting on Microsoft Fabric community.
The Monitor duration is the end-to-end wall-clock time (it includes waiting in the capacity queue, model warm-up...). The Refresh ID time shows only the actual processing once a slot was obtained (Power Query/partition processing + commit, plus query cache).
So if Monitor says 7 to 8 minutes but Refresh ID shows 1 to 2 minutes, the extra minutes are mostly queue or warm-up time because the capacity was busy or just cold.
It is better to use Monitor as the total user wait and Refresh ID as the engine compute time.
- Monitor duration
- includes total job time—like queuing, cache refresh, and post-processing.
- Refresh ID duration shows only the actual data refresh time (partition processing).
- That’s why monitor shows ~7 min, but refresh ID shows ~2 min.
Use monitor for capacity insights, and refresh ID for performance tuning. Let me know if you want a quick visual tracker for both!
That would be great, thank you!
Can I gather this type of information from the powerbi rest apis?
- Monitor duration
