The Future of Hazelcast
summary
Hazelcast moves from cache to a cloud-native platform: 5.2 ships with zero config serialization that eliminates uploading Java jars for ~80% of cases and adds SQL and zero-code map loaders; 5.5 merges Jet and IMDG into Hazelcast Platform so streaming state and maps are one product; Viridian offers an opinionated cloud with pay-by-data simplicity; a thread-per-core rewrite (Project Alto) prototypes show orders-of-magnitude latency and throughput improvements; a tiered storage preview enables terabyte-scale nodes with disk-backed cold data.
AI-written summary · about
chapters (11)
- 0:00Introduction and overview
- 2:35From cache to IMDG evolution
- 7:10Caching patterns and trade-offs
- 10:40Cloud and Kubernetes challenges
- 15:21Polyglot and zero-code initiatives
- 18:31Hazelcast Jet and streaming engine
- 21:01Platform merger and use cases
- 22:25Viridian cloud product
- 23:29Tiered storage and 6.0 updates
- 24:34Project Alto and future plans
- 28:24Q&A session
related talks
28mFlexible Cloud, Whatever The Workload
40mDeconstructing Apache Hadoop: A Dialogue About a Cloud-Native Refactor
16mHadoop's Impact on the Future of Data Management
43mWhere Is My Cache? Architectural Patterns for Caching ...
16mPreview: Cloud Computing
01h 03mDiscovering the value of a cloud data platform
15mPivotal CF, Everything Changes
32mHow we build a multi-AZ cloud in Switzerland