How big of a cluster can the kubernetes dashboard handle?

[msolberg8]

Hello,

We have a decent sized cluster that runs all of our non prod environments.

If you're wondering, we have:

41 nodes
1500 deployments
3200 pods
600 jobs
12k replica sets (considering droping the history down)

So a healthy sized cluster but still well within the supported range kubernetes.

My main question is how does the kubernetes dashboard perform with larger clusters.

Right now we're using rancher and the newer versions of rancher take loads of chrome memory to run since it seems to be loading everything into memory all the time.

Thanks in advance!

Michael

[msolberg8]

I've looked through some of the open/closed issues surrounding larger clusters and it seems like we'll be alright if we turn off / slow down auto refresh and jack up our mem limits.

I'd still like to hear any examples of success / failure with the dashboard in larger clusters though! Need all the ammo I can get to try and dump rancher.