What You Get by Replicating Lucene Indexes on the Data Grid
summary
Stores Lucene indexes as byte-array fragments inside an Infinispan Directory that replicates across the grid, keeps a single-writer master for commits, and reassembles segments on read; cache stores (Cassandra, HBase, S3) provide optional persistent backing. Implementation is a small Java patch with extensive concurrent tests, used to scale Jira indexes, and shows query/write behavior roughly between RAM and FS directories while allowing transactional index updates and linear scale by adding nodes.
AI-written summary · about
chapters (21)
- 0:00Introduction
- 1:11What will we talk about
- 1:50What is a partial
- 5:05In practice
- 5:45Distribute data
- 6:21JGroups
- 7:14Other Languages
- 8:06Consistent Hashing
- 9:21Transactions
- 10:10Cache Loader
- 12:09Single Machine Indexer
- 12:51Hibernate Search
- 13:59How does it work
- 14:42Size
- 15:37Master
- 16:33Example
- 16:59Thoughts
- 17:36Performance
- 20:42Future plans
- 21:40Conclusion
- 22:06Questions
related talks
54mThe JBoss Data Grid, or Enterprisegrade Infinispan
38mBoost Your Applications with Distributed Caches/Datagrid
44mBoosting your applications with distributed caches/datagrids
47mBoosting your applications with distributed caches/datagrids
30mCaching for Speed: Practical Performance Optimization in OpenSearch & Lucene
36mSearch on the fly: how to lighten your Big Data
43mCaching and Beyond - Smarter Data Processing with Java
41mDistributed Java DBs Under the Hood: Main Components and Interactions Between Them