August Newsletter - Cloudberry Database(2023/08)

[tuhaihe]

Hi, welcome to our Cloudberry Database project newsletter! From this month, we will initiate biweekly/monthly newsletters to gather information and new statuses around the Cloudberry Database.

Pull Requests

• [FIXME] Wrong results on main branch for INDF query by @wenchaozhang-123
• Doc: update security policy and PR template by @tuhaihe
• Catchup 1X_STABLE by @Ray-Eldath
• Add extensible smgr slot for other storage format. by @wenchaozhang-123
• Doc: update editor settings for CloudberryDB by @tuhaihe
• Doc: update the building guide for Linux by @Zhangbaowen-Hashdata
• Feature: add user-defined index access method by @hw118118
• Fix create ao table set parallel_workers doesn't take effect by @Ray-Eldath
• Feature: add user-defined index access method by @hw118118
• Support create/drop database in serverless architecture by @roseduan
• Fix: use RelFileNodeId instead of Oid for relfilenode. by @HuSen8891
• Adjust the method of loading the external preloaded shared libraries by @MisterRaindrop
• Change json_object name to avoid conflict with jansson lib. by @MisterRaindrop
• Fix unstable tests reindextable_while_reindex_idx by @Ray-Eldath
• Adjust the method of loading the external preloaded shared libraries by @MisterRaindrop
• Change json_object name to avoid conflict with jansson lib. by @MisterRaindrop
• Improve makeArrayTypeName's algorithm for choosing array type names. by @lss602726449
• Implement Parallel-aware Hash Left Anti Semi (Not-In) Join by @avamingli
• Fix bugs which could not create unique index "gp_fastsequence_objid_objmod_index" by @lss602726449
• Add final Motion earlier in grouping_planner for partial paths by @avamingli
• Fix: warehouse is not supported without unionstore by @roseduan
• Support separation of catalog and compute by @HuSen8891
• FIX: constraint check exception with snapshot memory leak in AO/AOCS by @lss602726449
• add hashdata extension storage_am related hook, by @wangliang03
• Loosen the restriction outer path has Motion of parallel plan by @avamingli

In addition, we also have updates for our Cloudberry Database website. Some new pages have been live, including security policy, contribution guide, working with Git & GitHub, code Contribution guide, proposal guide, and more. You can visit our website to learn more.

Proposals & Ideas

• [Proposal] Refactor ORCA to support user-defined access method in pg by @wfnuser
• [Ideas] Should prune tags after fetching from GP repo by @Ray-Eldath
• [Proposal] Implement Scorll Parallel Retrieve Cursor. by @wenchaozhang-123
• [Ideas] Wanna PostgresML in CBDB? An AI application database. LLM embeddings, generate text, or make online predictions using only SQL. by @avamingli
• [Proposal] Support Incremental View Maintenance (IVM) in Cloudberry Database by @avamingli

Issues

• Support dynamic table for Cloud by @my-ship-it
• ERROR:CBucket.cpp:45: Failed assertion by @liyxbeijing
• Upgrade pg_vector to 0.5.0 by @my-ship-it
• cdbd primary segment failure: writer proc reference shared with reader is invalid by @liyxbeijing
• SPIKE memory usage difference between PG and Cloudberry by @my-ship-it
• [Enhance] Too many log messages for memory stats when run out of memory by @gfphoenix78
• Support elastic search SSL for zombodb by @my-ship-it
• [Bug] gppkg -r is invalid by @chenfool

Contributors

🎈️🎊️ Thanks to the following contributors for helping make Cloudberry Database better this month:

@wenchaozhang-123, @tuhaihe, @Ray-Eldath, @Zhangbaowen-Hashdata, @hw118118, @roseduan, @HuSen8891, @MisterRaindrop, @lss602726449, @avamingli, @wangliang03, @my-ship-it, @liyxbeijing, @gfphoenix78, @chenfool, @TomShawn, @wfnuser

Stay updated

For more updates, please follow us on Twitter. If you have any questions on the Cloudberry Database, you can join our community channels to ask.

See you next!