SSTable Format Guide
SSTable Format Guide
Section titled “SSTable Format Guide”A 22-chapter reference for the Apache Cassandra 5.0 SSTable binary format, audited against Cassandra 5.0.8 as part of CQLite epic #598.
Source of truth:
docs/sstables-definitive-guide/— pages below are generated at build time bywebsite/scripts/sync-format-guide.mjs.
Chapters
Section titled “Chapters”- What Are SSTables?
- Anatomy of an SSTable
- Disk and IO Model
- From CQL to Disk
- Data.db Format
- Index.db and Summary.db
- Filter.db (Bloom)
- Statistics.db
- CompressionInfo.db and Chunking
- Point Reads and Slices
- Merging, Tombstones, and Shadowing
- Caching and OS Interaction
- Secondary Indexes (2i)
- Storage-Attached Index (SAI)
- Compaction Strategies
- SSTable Lifecycle and Maintenance
- BTI (Big Trie-Indexed) Formats
- Repair, Streaming, and Bootstrap (Overview)
- Example reference listing sourced from canonical datasets (names illustrative)
- Checksums and Integrity
- How to find a row by key (flow card)
- Versioning and Format Matrix (quick reference)
Appendices
Section titled “Appendices”- Appendix A — CQL->SSTable Type Mapping
- Appendix B — On-Disk Encodings Cheat Sheet
- Appendix C — Reference Walkthroughs with Code
- Appendix D — Tools & Workflows
- Appendix E — Glossary
- Appendix F — Known Limitations
- Appendix G: Cassandra 5.0 Compression Chunk Formats
- Appendix G: Algorithm Reference Sheet
- Appendix G Quick Reference - Compression Formats at a Glance
- Appendix H — Cassandra 5.0 CommitLog Segment Format