In-Memory Computing Meets Kubernetes
Kubernetes has become the de facto standard for containerized application deployment. Kubernetes can be used in any type of environment–on-premise, cloud, or hybrid–and is supported by every major cloud providers. In this session Yoav will focus on how to deploy multi-region data centers with Kubernetes clusters for real-time access to live operational data at sub-second latencies. Data stored in-memory is replicated asynchronously and no database is required to replicate data over the WAN. He will also cover the logic of tiered storage and how to implement it on Kubernetes to optimize access to SSDs and Storage Class Memory – allowing a hybrid storage model where data is stored according to the priority set by the application. By customizing hot data selection, the importance of data is dynamically controlled to utilize RAM for critical while saving (evicting) the desired data to multi-tiered data storage layers.