Performance comment ??

[hayden-t]

In the docs here: https://wrench.readthedocs.io/en/latest/performance.html
You say
"Wrench uses a single-process server, without threads, and blocks while processing data from any client. This means it has little hope of scaling in production."
What would be the capacity ? I guess relative to data size ?