Afi SV’s Post

Here are 3 types of scan you didnt know about in postgres Sequential Scan Index Scan & Bitmap Heap Scan Use the EXPLAIN or EXPLAIN ANALYZE infront of your Query Outputs the type of scan, the planning time and the execution time

Demo on working of indexes in postgres - how creating index on a field changed postgres to use index scan instead of parallel sequence scan. Follow my content here https://lnkd.in/gpF-8-KJ on a discussion around this.

💡 Faster B-tree index scans in Postgres 17 for queries that involve IN lists or other cases where multiple array values are being passed to Postgres... - https://lnkd.in/dGzDNnSm

Postgres 17 is released with great performance features! In this video I discuss one of the major performance features with BTree scan optimization when using IN with constant values. In prior releases, postgres would loop and do multiple "primitive" index scans as many as your IN elements. In Postgres 17 it takes an array and eliminates the need for multiple scans. Watch the full video https://lnkd.in/gKv6_EM8

Postgres 17 is released with great performance features! In this video I discuss one of the major performance features with BTree scan optimization when using IN with constant values. In prior releases, postgres would loop and do multiple "primitive" index scans as many as your IN elements. In Postgres 17 it takes an array and eliminates the need for multiple scans. Watch the full video https://lnkd.in/gKv6_EM8

You probably don't need a vector database. Postgres is still all you need My recent blog post spurred some interesting discussions around databases on HN and Reddit 🐘 Read about working with vector embeddings in Postgres, the launch of pgvectorscale and benefits of colocating data. If you're working with vector embeddings or considering using a vector database, this post is for you. Link to full post in comments 👇 Let me know what you think #postgres

Spent some time today going through the source code of PostgreSQL to understand its query parser ⚡ I compiled my learnings in a video along with a code walkthrough covering interesting internal details about what happens when a Postgres server receives a SQL query from the client, and how it parses the query to create the parse tree. give it a watch - youtu.be/m8PtOY3aWL0 ⚡ I keep writing and sharing my practical experience and learnings every day, so if you resonate then follow along. I keep it no fluff. youtube.com/c/ArpitBhayani #AsliEngineering #PostgreSQLInternals #DatabaseEngineering

Postgres stores `TIMESTAMP WITH TIME ZONE` columns converted to UTC, and it does not preserve timezone information. This can lead to nasty bugs! Simon Willison wrote an excellent blog post on how to store it properly: https://lnkd.in/gNdGWWtK

How to Remove an Orphan Large Object in PostgreSQL With Vacuumlo https://lnkd.in/eBeR2ryy

This PostgreSQL extension makes your vector search blazingly fast. ⚡️ You may know that you can use PostgreSQL with the PGVector extension to perform vector search. But it's not as fast as its competitors like Pinecone or Qdrant. PGVectorscale, written in Rust, aims for super-fast vector search for Postgres with PGVector. Look at the comparison below where Postgres gets a huge performance boost. Check it out: https://lnkd.in/dw2QqgsK