Benchmarks!

Should be pretty easy to throw together some benchmarks comparing the different async variants against their synchronous counterparts to see how much overhead the async state machines introduce.

Start by locating the async variants and their synchronous counterparts in the repository, then inspect any existing benchmark or test setup. Add comparisons for each matching pair and report the overhead introduced by the async state machines; done means the variants are benchmarked against their synchronous counterparts.

Written by the indexing model from the issue text.