We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
What Timescale's Approach to Dropping Chunks Tells Us About Concurrent DDL: Christopher Travers
Explore the complexities of concurrent DDL operations in PostgreSQL, including Timescale's approach to dropping chunks, and learn how to overcome performance and visibility issues with partman and background workers.
- Concurrent DDL operations are complex and can lead to problems
- Timescale’s approach to dropping chunks tells us a lot about concurrent DDL
- Fixing how we do relcache could solve some problems
- Visibility mappings can be a problem in concurrent DDL
- Currently, concurrent DDL operations can lead to poor performance and visibility issues
- Fixing these issues is non-trivial and requires careful consideration
- Some problems cannot be solved, only improved
- Partman can help with concurrent DDL operations
- Transparent concurrent operations are beneficial, but not always possible
- Real problems vs. fake problems: we should focus on real problems, not theoretical ones
- State diagrams are important in concurrent DDL operations
- Locking can be problematic in concurrent DDL operations
- Background workers can help with concurrent DDL operations
- Asynchronous cleanup can reduce complexity in concurrent DDL operations