Vertical scaling of HAProxy instances

Hello,

We are trying to vertically scale our HAProxy instances, and we are not getting the results that one would expect by upgrading the hardware (assuming that the software can take advantage of the extra resources).

We upgraded from machines with 16 threads, to machines with 32 threads, and we are only observing a 50% increase in the ability to sustain connections and rps, as well as ssl rate, and we can’t seem to reach that rate before we overload the server.

Custom kernel parameters (click for more details) HAProxy config (click for more details)

I’ve recently posted about “Theoretical limits for a HAProxy instance”, where I used the "Small" server as an example for the limits we were observing. I am using the same metrics here.

"Small" server specs (click for more details) "Small" server Prometheus metrics (click for more details)

This is the same test, in a bigger server with production traffic, but raising the maxsslrate and maxconnrate, from 2500 to 5000.

"Big" server specs (click for more details) "Big" server Prometheus metrics (click for more details)

We are wondering if:

1. Are these results expected?
2. Does anyone with a similar setup/config get different results?

2 posts - 2 participants

Read full topic