Hi Paul, I don’t think you’ll get anything other than duration, start and end date time. You could look at duration and compare it to the difference between the start date time and the end date time. If could be that duration purely relates to database refresh - but if you think in terms of instance details you get a create time, start time and end time plus duration…and the start time vs end time is the same as duration - so you can’t tell the difference between a report that took 90s on the database and 10s to distribute and a report that took 10s on the database and 90s to distribute.
That’s pretty much the conclusion I’d come to.
What’s behind the question is that the number of reports has outgrown our present hardware, and although we are upgrading this, there is a lot of talk about “demand management”.
This question came up as one way to determine the most “expensive” reports
There are some different Event Types though. You might be able to break some information out by those. For example, I would assume that the Refresh event type is for retrieving data from the database and the Retrieve event type would be for retrieving the report from the FileStore. If I remember correctly, it may be hard to correlate these events to a single instance of a report running because they have different Event IDs.
This is an interesting topic but I don’t have time to dig into it right now. Wish I did.