“JRuby file IO using mmap”

This is my quest for achieving fast disk IO for a JRuby persistent queue implementation. This talk will present the possible strategies for doing disk IO, the pitfals of using Java objects in JRuby for data intensive code, and how to write a Java extension to avoid the high cost of type conversion when crossing the world between JRuby and Java. This talk will be a deep dive into JRuby and Java code with benchmarks for all strategies.