Rethinking Microservice Persistence: Beyond the Database Monolith
summary
Argues that a Java-native persistence layer removes the “database black hole” by storing and querying live object graphs in memory with Eclipse Store and the Eclipse serializer, backed by OpenJ9 and an open-source Eclipse data grid. Shows automated indexing and lazy loading (gig mapap), claims microsecond queries up to 1000x faster than DBs, and describes a distributed micro-database network and micro vector database that spins up tiny, serverless data nodes for isolated, rapid recovery.
AI-written summary · about
chapters (15)
- 0:00Introduction and welcome
- 4:25Eclipse Foundation overview
- 7:17Introducing Eclipse Store
- 11:52Indexing with GigMap
- 13:01Distributed Eclipse Store
- 15:06Microservices with Eclipse Data Grid
- 17:39Rethinking database architecture
- 21:01Future of micro databases
- 22:12Transition to Mark Studley
- 22:33OpenJ9 and Semeru runtimes
- 28:11Performance evolution of OpenJ9
- 31:22InstantOn technology
- 33:07Cloud compiler and JIT server
- 34:51Memory optimizations
- 37:17Downloading and support
related talks
40mHow to Build Ultra-fast Java Microservices. Overcome Pitfalls of Data Persistence
43mThe Java Cloud-Native Stack for Microservices and Serverless Architecture
01h 00mEclipse MicroStream - Ultra-fast Java cloud-native persistence for microservices
55mDatabaseless High-Performance Persistence with EclipseStore...
43mEclipseStore - Ultra-Fast Java Cloud-Native Persistence for Microservices and Serverless
56mEclipseStore - Databaseless Java Cloud Native Persistence for Microservices
52mGet Started Coding with EclipseStore...
22mEclipseStore – The new, ultra-fast Java persistence for the cloud
