Fork me on GitHub
#proletarian
<
2021-08-04
>
msolli07:08:05

Thanks, @dominicm, that’s useful information. It would be nice to have a benchmark in the repo to establish a baseline for performance. Is this something you could share (in a gist or something)?

dominicm19:08:30

@msolli It was pretty basic, I was technically testing a bunch of other stuff too (e.g. our auto-scaling setup, our http layer, my local network). So it's not a great test at all. For my test, I used wrk with a lua script to set the HTTP body.

dominicm19:08:59

I suppose the processing is more interesting, but for that I was measuring with our datadog setup. So, again, not great for a gist.

dominicm19:08:20

Maybe something could be put together with https://github.com/aphyr/interval-metrics#measuring-your-codes-performance You'll still want to externalize the PG instance somehow, so the JVM doesn't impact it and it's an accurate measure of the latency you will see in production.

👍 5