📚 New: Apache Cloudberry Core Internals is live!
A deep-dive handbook on how Cloudberry works inside:
https://cloudberry.apache.org/core-internals/
What’s inside (21 chapters):
• The big picture: MPP architecture, QD/QE, dispatch, gangs and slices
• Storage: heap, AO/AOCS, PAX, buffer cache, indexes, catalog
• Transactions: MVCC, vacuum, global deadlock detector, distributed 2PC
• Query processing: planner, ORCA, executor, Motion and interconnect
• HA and recovery (WAL, replication, FTS), plus resource groups
Huge thanks to
@Max Yang for writing and contributing it! 🙏
If you’re new to the codebase, Chapter 1 has a module map that’s a great place to start. Feedback and fixes are welcome as PRs to
https://github.com/apache/cloudberry/tree/main/doc/core-internals |
doc/core-internals
in the main repo.