Performance of createListenerMiddleware Predicate in Redux Toolkit

[guyeilon]

Hi Redux Team,

I have a question regarding the createListenerMiddleware in Redux Toolkit. Specifically, I want to know how the predicate function behaves in terms of performance.

My Question:

• Does the predicate function get evaluated after every action is dispatched, regardless of the action type or target?
• If so, could this cause performance issues in applications where actions are dispatched frequently, particularly if the predicate contains more complex logic (e.g., state comparisons or function calls)?
• Are there any best practices or recommendations to optimize the performance of createListenerMiddleware when using predicate?

I’m concerned about whether using predicate in a heavily action-driven environment could degrade performance, and I’d appreciate any insights or suggestions to avoid potential bottlenecks.

Thanks in advance for your help!

[markerikson]

Yes, it does get evaluated after every action.

However, it shouldn't be a perf bottleneck. Predicates are supposed to be trivial checks that immediately return a boolean. That's even less work than all the selectors that will run after every action.

Now, yes, those are more function calls, so there is an O(n) aspect to perf here - 100 predicate calls is literally more expensive than 50 or 10 or 1 predicate calls. But in practice these should have effectively no meaningful impact on perf, especially compared to rendering components or anything else.