Netflix created and open sourced Dynomite project to provide reusable distributed database infrastructure that turns single server data stores into scalable, distributed databases. Dynomite supports pluggable protocols and pluggable storage engines, which allows us to add sharding and replication to a variety of non-distributed data stores. The entire database infrastructure can be reused across a variety of workloads from in-memory to on-disk, and across APIs from key/value to document databases. Dynomite allows application developers to choose the API that best fits their requirements, while DevOps can select the best operation database based on the workload. Dynomite is used by Netflix to handle millions of OPS in production leveraging Redis and RocksDB. In this talk, we are going to show how we achieved high availability by being able to terminate any Dynomite node without client side downtime, best practices and the challenges in deploying Dynomite in production.
Ioannis Papapanagiotou is a senior distributed systems engineer at Netflix's Cloud Database Engineering team and a graduate faculty at Purdue University. Ioannis has a dual PhD in Computer Engineering and Operations. He has served in the faculty ranks of Purdue University, University of New Mexico, and NC State University. He has been awarded the NetApp faculty fellowship and established an Nvidia CUDA Research Center at Purdue University. Ioannis has also received the IBM and Academy of Athens PhD Fellowships for his PhD research, teaching awards, and best paper awards in several IEEE conferences for his academic contributions. Finally, he is a member of ACM and senior member of IEEE.
Content Curator: Dini Martini