asked on June 17, 2025
I'm currently implementing a custom API,
I've noticed in Flex Monster (f/e) is continuously calling pages increasingly even if user doesn't take any action (such as scrolling).
The size of our data is so big so we apply pagenation with custom api,
but it takes too long to draw a grid after receiving the entire data.
Can I add a page number and send an api call only after I scroll through it?
⋅ June 18, 2025
Hello,
Thank you for reaching out to us.
The current workflow with all pages loaded consecutively is intended and cannot be changed. In the current implementation, the pagination only changes the request handling logic on the server. At the same time, the UI would still wait for an entire dataset to load before rendering.
However, we are working on a new major version of Flexmonster that would work similarly to what you described. After loading a data block, it would be immediately rendered in the table without waiting for all data outside the visible range to be loaded, and scrolling would trigger a request to get a new chunk of data. While the performance is one of the biggest priorities in developing a new version of Flexmonster, it also has other significant architectural changes. You can find more information about Flexmonster 3.0 in our blog post: https://www.flexmonster.com/blog/flexmonster-3-0/release-3-0-roadmap-from-cto/
Please let us know if any further questions arise.
Best regards,
Maksym
⋅ June 23, 2025
Hello, thanks for the answer. So does that mean the scrolling feature is currently unavailable? If it's under development, I wonder when it will be released.
⋅ June 23, 2025
Hello,
Thank you for your reply.
Yes, the data loading on scroll is currently unavailable. This feature is just a small part of what is coming in Flexmonster 3.0 - a complete rewrite designed for handling larger datasets and a more flexible, dashboard-focused architecture. Our team will contact you when a prerelease version is available. Please note that the prerelease will be presented as an interactive demo only and won't be available for download at this stage.
Best regards,
Maksym
⋅ July 1, 2025
Can I know the 3.0 deployment date? I'm asking because our system is short on schedule.
⋅ July 1, 2025
If it is difficult to use right now, I would appreciate it if you could tell me how to customize the f/e code.
⋅ July 1, 2025
https://www.flexmonster.com/question/very-very-huge-big-pivot-table/
According to the previous response about custom API continuously loading all pages until the entire dataset is loaded before rendering, doesn't this mean that all data still needs to be stored in browser RAM, which creates memory limitations?
If custom API loads all the data anyway (just in chunks rather than all at once), what's the actual benefit of using it? If it's going to load all visible data regardless, I don't see the point.
Is there any way to implement true pagination functionality in the current version where data is only loaded when needed (e.g., on scroll)?
⋅ July 1, 2025
Hello, Taehong and Jhshin!
Thank you for getting back to us.
We fully understand that for huge amounts of data, loading data in chunks based on scroll position can be an effective approach. However, this behavior is not part of the current design. At the same time, our team is paying attention to clients' needs and expectations, so we are already working on a completely new version of the component, where the data chunks would be loaded only when the user scrolls to the needed area.
We are preparing to launch a preview of the new 3.0 version soon, however, it will be ready for beta testing later in the year. Flexmonster 2.9 remains our stable production version, and our team recommends continuing to use the current version and its features in your development.
Regarding your question about the benefits of custom data source API protocol, please note that the main advantage of using our communication protocol is that only aggregated data is loaded to the client side and is stored in the browser RAM. The actual size of data loaded into the client depends on the number of fields, unique members, and measures, but for the pivot table, it should be significantly smaller than the original data size. It is important to note that on the flat table, Flexmonster will load all the raw data to the client, and using the custom data source API won't benefit much.
Flexmonster itself does not have limitations for the data size, but if the size of aggregated data is still bigger than the browser can handle, we recommend handling more complex calculations on the server, prefiltering data on the grid, and using reports with smaller data slices for your analysis.
Hope we have addressed your questions.
Let us know if there's anything else we can help with.
Kind regards,
Solomiia
