#528 — November 1, 2023 |
|
Postgres Weekly |
Making Postgres Tick: New features in Marco Slot (Citus Data) |
Auto Optimize Your RDS or Aurora Postgres Databases With AI — Need help optimizing Postgres knobs, queries, indexes and autovacuum? OtterTune is your AI-powered database co-pilot that can improve performance, reduce costs and keep your databases healthy. Get a free, 30-day trial for three databases. OtterTune sponsor |
Nile: A New Serverless Postgres Platform — Neon is perhaps the best known serverless Postgres platform right now, but some fresh competition is on the way. Nile’s angle is a focus on SaaS usage, supporting multi-tenant setups with seamless tenant isolation and scaling. It’s behind a waitlist, but this post goes into technical depth on their approach. Sriram Subramanian (Nile) |
🗣 Nile's announcement provoked extensive discussion on Hacker News oriented heavily around the pros and cons of their multi-tenant focus. |
SQL Identity Columns — You might also call them autoincrementing, or serial columns, but such identity columns were (eventually) standardized in SQL:2003 and Postgres has supported their Peter Eisentraut |
IN BRIEF:
|
▶ Storing and Querying LLM Embeddings with Django, Postgres and pgvector — We’ve mentioned pgvector a lot recently, and here’s a look at a practical use case for it – building an LLM embeddings based search system that enables searching over job descriptions using an unstructured, English-language description of a potential job seeker. GitHub repo. William Huster |
GDB Debugging Essentials with Postgres — Explore the essentials of debugging with GDB, using Postgres as a practical example. (Note: This is not needed by day-to-day Postgres users, but if you’re a C/C++ developer or working on Postgres internals, you’ll find it useful.) Cary Huang |
📅 Webinar on Nov 15: Automating Index Selection Using Constraint Programming pganalyze sponsor |
What Is TOAST (and Why It Isn’t Enough for Data Compression in Postgres)
|
How to Understand Explain Plans
|
Code and Tools |
PL/Haskell 3.1: Use Haskell in Your SQL Functions — Haskell is a functional language but you can, nonetheless, use it to define procedures in such a way that Postgres can play nicely with. Edward F. Behn, Jr. |
pg_graphql 1.4.2: Bring GraphQL Support to Postgres — Reflects a GraphQL schema from the existing SQL schema, letting any GraphQL client query the database via GraphQL with no additional servers, processes, or libraries. Supabase |
|