An Algorithm for Incremental Backups in Apache Ignite
summary
Describes an algorithm for creating incremental backups (snapshots) in the distributed in-memory database Apache Ignite, addressing the additional challenges introduced by distribution and support for distributed transactions. Explains how the incremental approach works to capture changes efficiently across cluster nodes while preserving transactional consistency.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:10About Apache Ignite
- 3:14Full backups problem
- 6:05Need for incremental backups
- 8:52Storing operation history
- 10:17Finding incremental backup algorithm
- 14:30Algorithm based on partially ordered events
- 15:56Algorithm metaphor with stars
- 20:11State capture and storage optimization
- 23:25Implementation in Apache Ignite
- 28:02Results and benefits
- 29:31Implementation challenges
- 30:36Q&A
related talks
56mHow to Take a Backup in a Distributed System Without Anyone Noticing
49mBeyond the Data Grid: Fast Data Processing with Apache Ignite
49mIncremental Backup
39mIncremental Backups in PostgreSQL
14mIncremental snapshots/backups in KVM plugin of Apache CloudStack
59mApache Ignite Persistence
18mIncremental Backup in PostgreSQL
49mIncremental Computation