

Bitemporal Databases: What They Are and Why They Matter, with James Henderson of XTDB
9 snips Jun 7, 2023
James Henderson, lead developer of XTDB, shares insights on the intriguing world of bitemporal databases. He discusses XTDB's ability to manage two notions of time, crucial for sectors like accounting and auditing. The conversation dives into architectural challenges, query optimization with Datalog, and the advantages of using Clojure for rapid development. Listeners learn about hybrid databases and the role of community feedback in innovative database technology, making it a must-listen for developers interested in cutting-edge database solutions.
AI Snips
Chapters
Transcript
Episode notes
Understanding Bitemporal Data
- Bitemporality tracks both valid time and transaction time for data changes.
- This enables retroactive and proactive updates and querying data as known then or corrected now.
Inside-Out Log Architecture
- XTDB uses an inside-out architecture centering the transaction log as the data's source of truth.
- The database state is a projection derived from this immutable log, similar to event sourcing.
Managing Immutable Logs Safely
- Be conservative when making changes to an immutable transaction log.
- Use separate indexing structures for new features to maintain backward compatibility.