Jens Wilke

When doing benchmarks we are usually interested in comparing the throughput of different solutions. However, when the code under investigation is not trivial, memory consumption may vary significantly, because using more memory to achieve better throughput is a typical design trade off.

The talk discusses what memory metrics are more or less meaningful and what approaches can be used to track memory consumption alongside JMH benchmarks. A special focus will be given on dynamic effects on the interaction with the garbage collector and whether it is possible to get some stable benchmark results.

The development of cache2k has two streams: It is a robust cache implementation and provides features we miss from other products, on the other hand it is a research area to try out new caching algorithms and implementations to achieve the best performing in-memory Java cache that is available today.

The talk will start with some real-world scenarios and challenges and how the cache can help, making it the pivotal point in an application to manage data resources and provide them on time. The second part of the talk will be about modern and adaptive caching algorithms and why you should not implement yet another "LRU cache" by yourself.

This talk should encourage to use more caching to speed up applications and, of course, to hack and improve existing caching products. There is still much to achieve!