An Introduction to Bε-trees and Write-Optimization
Web page · https://usenix.org/publications/login/oct15/bender
This is the underlying tech of the recently posted GEFS: https://lobste.rs/s/zjjvbo/work_on_gefs_good_enough_file_system_for
First seen by Oriel on .
AI summary
This article introduces the Bε-tree, a write-optimized key-value data structure that can outperform B-trees, particularly for inserts, range queries, and key-value updates.
- Bε-trees can organize on-disk storage for applications such as databases and file systems.
- The article compares the asymptotic performance of Bε-trees with B-trees and Log-Structured Merge trees (LSM-trees).
- Real-world performance measurements complement the article’s theoretical comparisons.
- Readers learn how Bε-trees work, their performance characteristics, and how to design applications to benefit from them.
#bε-trees #data structures #storage systems #databases
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Points
Score reported by the source.
- Lobsters: 26, latest observation ; +23 over observed period since
Comments
Comment count reported by the source.
- Lobsters: 1, latest observation ; +1 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| Lobsters | 0ec9yk | https://www.usenix.org/publications/login/oct15/bender Read discussion |