This project was created to microbenchmark the efficiency of different Set and Map implementations. In particular, HashSet versus EnumSet, and HashMap versus EnumMap
These are very simple benchmarks, and at present only create each collection, and perform a few common operations on them.
Use is made of https://github.com/openjdk/jmh.
This is a synthetic benchmark, and real world performance will inevitably be different.