Description:
We are using Flexmonster version 2.9.43 embedded in a JavaScript web application with a JSON inline data source.
Issue: When loading a report with a large dataset and a layout that has multiple row dimensions and several active measures, Flexmonster displays the following message and prevents expanding nodes:
"The dataset is too large. Some fields can't be expanded. Please narrow down the dataset."
Steps to reproduce:
3. Load the report — the error message appears and nodes cannot be expanded
Analysis: Even at the leaf cell level alone, before any subtotals are counted, the total number of cells required already exceeds what appears to be Flexmonster's internal rendering limit:
~30,000+ unique values (high-cardinality row dimension) × ~18 column values × 8 active measures
= ~4–5 million leaf cells
Questions for Flexmonster support:
Environment:
Hello,
Thank you for writing to us.
Expanding all nodes with a high-cardinality dimension in rows and many measures is a heavy operation, so hitting this message is expected in your scenario.
To answer your questions:
expandExecutionTimeout option. You can increase it, but keep in mind users will have to wait longer for expands to complete.expandAll.Please let us know if our answer was helpful.
Best regards,
Maksym
Hello,
Hope you are doing well.
We are wondering if you have time to review our recommendations and if increasing expandExecutionTimeout, or pre-filtering the dataset resolved the issue.
Looking forward to hearing from you.
Best regards,
Maksym
Mksym,
Thank you for following up.
We increased the timeout but still it after a number of cells in our expanded report it stops. Somehow my team found that 500-1000 million of cells is the mas will expand.
Sure filtering works, and not expanding all data at once works. So at this point expanding by the higher group works but ..... This is not the requirement, the customer is quite upset that they cannot expand all and have a full view of their data in expanded mode where they can navigate quickly.
I used the data server and that seems to make it fine but, we don't usually provide that with our application so it is only a future option.
Honestly, with the amount of cells it is reasonable to wait. We are ok displaying a message about the wait while it expands.
The browser seems to cope well with the data, memory is ok.
I can provide a dataset and the save report so you can run the analysis and see if this is a feature you want to support. Do you want it? I can do it by tomorrow.
Unfortunately for me, I'm currently asking them to use excel for that expanded dataset/calculation.