Add benchmarks

[insanitybit]

It would be interesting to do a couple of benchmarks. In particular:

• equivalent Tokio / Futures implementation
• equivalent Erlang implementation
• equivalent Akka implementation
• equivalent Pony implementation

In particular, I want to see the difference between using futures and actors.

[insanitybit]

It would also be nice to have benchmarks so that issues that may improve performance (like the queue change) have something to go off of.