The State of Caching in Go

Aman will then present the design, performance and hit ratio comparison for the existing popular cache implementations in the Go ecosystem. After our failed attempt with lock striping with map shards, we modified this LRU cache by introducing locks and making it concurrent. Caffeine meets all the five requirements I mentioned at the beginning, so we are looking into building a Go equivalent of Caffeine, something which can serve our needs and potentially fill the gap of a concurrent, performant, memory-bounded cache in the Go language.

Source: blog.dgraph.io