Apache Ignite Persistence
summary
Apache Ignite now adds native disk storage to the in-memory platform so SQL queries and distributed computations work over the full dataset (not just cached memory), scaling to thousands of nodes and terabyte‑to‑petabyte class sizes. It implements page‑based B+‑trees, small page sizes to avoid hot/cold waste, configurable memory regions per table, a per‑node WAL + checkpoints recovery model, and uses H2 as the SQL planner; durability is synchronous by default but can be relaxed for latency.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:09Modern storage problems
- 3:57Need for horizontal scaling
- 7:07NoSQL emergence and limits
- 10:18Apache Ignite as solution
- 12:06Apache Ignite with SQL
- 13:10Data caching problems
- 15:39Need for disk integration
- 17:24Apache Ignite with disk storage
- 18:30Storage architecture changes
- 20:16Page-level data organization
- 23:48Memory tuning for regions
- 27:02Key-based query processing
- 31:58Horizontal scalability
- 34:07SQL query execution
- 36:14Summary and outlook
- 37:44Q&A
related talks
33mDistributed Database DevOps Dilemmas? Kubernetes to the rescue!
49mBeyond the Data Grid: Fast Data Processing with Apache Ignite
41mDistributed Java DBs Under the Hood: Main Components and Interactions Between Them
28mAerospike's Journey Through Innovations, Sustainability, and Roadmap
32mJava's Ignored Potential: Supercharge Database Apps with Java In-Memory Process
16mAerospike + Intel® persistent memory
53mApache Cassandra. Inception - all you need to know
45mHow to write fast in Apache Ignite