Hi team,
I’m currently building an enterprise dashboard application using Angular with EJ2 components (Grid, Charts, and DataManager). The project now requires integrating an edge AI inference layer that runs locally to generate real-time predictions based on user-selected data.
The grid displays operational datasets, and when a user selects specific rows, we pass that data to an edge AI model running on a local service. The returned predictions need to dynamically update multiple charts and summary panels without refreshing the entire page.
While basic binding works fine, we’re noticing minor UI lag when predictions are triggered frequently. I’d appreciate guidance on:
-
Recommended state management patterns for handling rapid prediction updates.
-
Whether batching updates performs better than incremental binding.
-
Any performance optimization tips for large datasets in real-time scenarios?
If anyone has implemented something similar, sample approaches or architectural suggestions would be extremely helpful. Thanks in advance!
