Skip to content
Notifications
Clear all

Anyone else having UI lag when loading reports with 1000+ items?

4 Posts
4 Users
0 Reactions
19 Views
(@metric_maverick)
Eminent Member
Joined: 7 months ago
Posts: 26
Topic starter   [#5152]

Just hit a major performance wall. UI becomes unusable when loading any report with a large dataset.

My benchmarks on a dashboard with 1,200+ items:
* Initial load: 12-14 seconds.
* Any filter/sort interaction: 5-7 second freeze.
* Browser memory usage spikes to ~1.8GB.

Tested on Chrome/Edge, different machines. Our smaller reports (<500 items) are fine. This seems like a frontend rendering bottleneck, not a data fetch issue—network tab shows data is delivered in ~2s.

Is this a known scaling limit? Any workarounds besides pagination (which breaks some required views)? Need hard numbers on what others are seeing.


Show me the numbers.


   
Quote
(@brian7)
Reputable Member
Joined: 3 months ago
Posts: 254
 

Yeah, same issue here with our analytics dashboards. We also see the data come in fast but the browser just chokes trying to render it all.

I'm wondering, is the lag happening during the actual DOM painting, or when the JavaScript is building the data objects? The memory spike makes me think it's holding everything in memory for the UI state.

Have you tried profiling it with the browser's performance tool? Might show which script or paint event is taking the longest.



   
ReplyQuote
(@chrisl)
Estimable Member
Joined: 3 months ago
Posts: 149
 

The memory spike typically confirms it's both: building large virtual DOM trees and holding full datasets in component state. The performance tool will show Recalculate Style/Layout as the main cost.

In our tracing, we found 70-80% of the lag was in repeated style calculations, not the initial JavaScript execution. Rendering engines recalculate even hidden elements unless you implement virtualization at the data layer.

A quick test: if you set `content-visibility: auto` on the report container, does the interaction lag drop? It won't fix memory, but it isolates painting cost.



   
ReplyQuote
(@alexh99)
Estimable Member
Joined: 3 months ago
Posts: 119
 

Those benchmarks look similar to what I've seen, especially the memory spike. The 2-second data fetch confirms it's not the database.

Have you checked if your visualization tool is trying to render every single item as a distinct DOM element, like individual chart points or table rows? That's usually the culprit at 1000+ items. Sometimes the tool defaults to a detailed view even when it shouldn't.

What are you using to build the reports? I've had this with some Looker explores when the visualization setting is wrong.



   
ReplyQuote