Sébastien Ros

I am talking about the request payload, not the response.

> so likely ~7.2/7.3m request per second The same initial conclusion we got to a few years ago. (sorry I forgot about bits/bytes) Which is why I am skeptical about...

There is one thing that is different for faf. I assume most frameworks use port 8080, but faf is using 8089. @errantmind would you like to submit a PR that...

Does validation run on CI runs? Because if external dependencies can be updated without submitting a PR we might miss some regressions. (floating dependencies, local GitHub repos clones)

An option we have tried successfully is to use the database server as a second load generator fort plaintext and json. It requires synchronization but it works, and showed that...

@talawahtech The wrk limitation on plaintext is probably coming from the fact that pipelining is handled by a LUA script. It its early versions this was a feature that was...

Before lua support was added to wrk, it could do pipelining. Take a look at these PRs: https://github.com/wg/wrk/pull/57 https://github.com/wg/wrk/pull/36

We have a jobs queue of both machines, and a "driver" app that orchestrate these, once the two jobs are ready to start, we send the command simultaneously, then aggregate...

At MS we run all the TE benchmarks with `--network host` for the same reasons.

> The benchmarks show best performing plaintext on physical to be at 2.7M rps. I can see 9M RPS with the latest runs on Citrine ... how is that possible...