I don't think the power of elixir sits in the benchmarks of this sort. Elixir is not a fast language.

Maybe a better benchmark is to compare the time it takes to setup, test, and run a distributed benchmark all in all 3 + elixir.