Skip to content

[Feature]: ? how to scale our local installation #2548

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
tuehlarsen opened this issue Apr 9, 2025 · 0 comments
Open

[Feature]: ? how to scale our local installation #2548

tuehlarsen opened this issue Apr 9, 2025 · 0 comments
Labels
question Further information is requested, label should be removed once answered

Comments

@tuehlarsen
Copy link

What change would you like to see?

Can you give some advises about how to scale our production installation or how to optimize it?

Context

We use k3s as cluster SW and have currently two nodes running in our local production setup.

The master which is virtualized - 32GB RAM, 8 CPUs, it is both master and runs workloads.
We have added the first physical worker node - 64GB RAM 24 CPUs, which only runs workloads and plan to change the master also to a similar physical and add more physical workload servers.

we have using the API to schedule 3 x 800 frontpage jobs without any problems. The last API schedule
was with one hop out. We have in the GUI in the last schedule seen 17 concurrently running jobs. The first 2 schedules took about 2 hours and
we did not see any running jobs in the GUI probaly because they ran so fast that it was not measured in the Dashboard.
We have seen 154 concurrent crawler pods and 478 pods in total during the crawls.

Here a top from the 2 servers:

The Master seems to be maxed out:
CPU %user %nice %system %iowait %steal %idle
11:40:00 all 47.22 11.20 32.08 0.08 0.00 9.43
11:50:01 all 49.38 11.07 31.85 0.07 0.00 7.63
12:00:02 all 46.75 11.14 33.12 0.08 0.00 8.91
12:10:00 all 49.52 10.85 33.26 0.07 0.00 6.30
12:20:00 all 48.00 11.10 32.16 0.08 0.00 8.67
12:30:00 all 50.42 10.78 30.90 0.33 0.00 7.56
12:40:00 all 48.26 11.13 31.63 0.10 0.00 8.88
12:50:00 all 50.89 11.06 30.58 0.13 0.00 7.34
13:00:02 all 49.20 11.07 30.84 0.10 0.00 8.79

The new workernode:
CPU %user %nice %system %iowait %steal %idle
10:40:00 all 0.40 0.00 0.19 0.00 0.00 99.40
10:50:01 all 0.23 0.00 0.17 0.02 0.00 99.59
11:00:00 all 0.36 0.01 0.19 0.00 0.00 99.44
11:10:00 all 18.58 0.00 9.22 0.00 0.00 72.19
11:20:01 all 13.12 0.00 6.82 0.00 0.00 80.06
11:30:00 all 27.21 0.01 12.76 0.08 0.00 59.94
11:40:00 all 28.86 0.01 13.53 0.03 0.00 57.58
11:50:00 all 18.24 0.01 12.78 0.01 0.00 68.97
12:00:00 all 16.38 0.01 12.51 0.01 0.00 71.09
12:10:01 all 19.25 0.01 12.83 0.02 0.00 67.89
12:20:00 all 18.49 0.01 12.96 0.01 0.00 68.53
12:30:00 all 28.97 0.02 13.79 0.04 0.00 57.18
12:40:00 all 23.93 0.01 13.81 0.04 0.00 62.22
12:50:00 all 33.36 0.03 16.04 0.31 0.00 50.26
13:00:00 all 27.66 0.01 15.37 0.04 0.00 56.92

@tuehlarsen tuehlarsen added the enhancement New feature or request label Apr 9, 2025
@SuaYoo SuaYoo added question Further information is requested, label should be removed once answered and removed enhancement New feature or request labels Apr 9, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested, label should be removed once answered
Projects
Status: Triage
Development

No branches or pull requests

2 participants