New caching algorithms performance evaluation

In this paper we propose new caching and test their performance. We use Memcached platform to test the caching algorithm performance. Memcached is an extremely popular open-source, distributed key-value store system that powers several of the Internet´s busiest websites such as LiveJournal, Wikipedia, Facebook and others. Currently, Memcached uses a trivial Least Recently Used algorithm for cache eviction decision-making. We describe modifications to the Memcached caching system and the introduction of alternate caching policies and benchmark them as well. 3 new caching algorithms are introduced and 2 new implementations of well-known algorithms are benchmarked as well. We discuss the caching algorithm and the simulation methods used in this paper. By introducing superior caching algorithms, we demonstrate a 10-20% improvement of hit rates. This feature has been sent to Memcached upstream and was met with positive feedback.