[PunchyHamster]

it's not really bottlenecked by the store but by the calculations performed on each pod schedule/creation.

It's basically "take global state of node load and capacity, pick where to schedule it", and I'd imagine probably not running in parallel coz that would be far harder to manage.

[senorrib]

No a k8s dev, but I feel like this is the answer. K8s isn't usually just scheduling pods round robin or at random. There's a lot of state to evaluate, and the problem of scheduling pods becomes an NP-hard problem similar to bin packing problem. I doubt the implementation tries to be optimal here, but it feels a computationally heavy problem.

[OvervCW]

In what way is it NP-hard? From what I can gather it just eliminates nodes where the pod wouldn't be allowed to run, calculates a score for each and then randomly selects one of the nodes that has the lowest score, so trivially parallelizable.

[femiagbabiaka]

I think filtering and scoring fall under a heuristics based approach to address NP-hardness?

Binpacking seems to be a well-defined NP-hard problem: https://en.wikipedia.org/wiki/Bin_packing_problem

[stevefan1999]

That's greedy

[__turbobrew__]

The k8s scheduler lets you tweak how many nodes to look at when scheduling a pod (percentage of nodes to score) so you can change how big “global state” is according to the scheduler algorithm.