https://itnext.io/tfservingcache-a-distributed-lru-cache-for-serving-tensorflow-models-in-kubernetes-d3c4bb959d56
