~postgresweekly | Issue 569 - The Tetris effect (10)

Tetris-SQL: Using SQL's Turing Completeness to Build Tetris — An implementation of Tetris in just a few SQL queries on Postgres! As well as being an impressive bit of work, it’s also educational, as Nuno takes time to walk through the implementation details in the README.

Postgres 17 Release Candidate 1 Released — If you’ve been uncertain about running any of the betas, now is a good time to kick the tires, as the release candidates tend to be largely identical to the final release. The final release is expected on September 26 (we may delay Postgres Weekly that week, due to this).

Run GitHub Actions Up to 2x Faster at Half the Cost — Blacksmith runs your GitHub Actions substantially faster by running them on modern gaming CPUs. Integrating Blacksmith is a one-line code change. 100+ companies like Ashby, Superblocks, and Slope use Blacksmith to help developers merge code faster.

Shrinking Big Postgres Tables: Copy-Swap-Drop — If you’ve got a big unwieldy table with only a subset of its data in active use but you don’t want to partition the table or perform lots of DELETE operations, you could copy the rows you want to keep into a new table, rename the new table, and drop the original one.

How Neon Dynamically Estimates Postgres’ Working Set Size — ‘Working set’ refers to the data that’s used/read/written by a system over a specific time period. Figuring out what the current ‘working set’ of a Postgres server is is actually quite difficult, but essential to do if you want to do autoscaling, as serverless platform Neon now can.

Understanding the Postgres Hackers Mailing List — If you want to get involved with Postgres development or otherwise have a deep interest in following along with developments, the pgsql-hackers list is the place to be. As a busy, highly technical list, there’s a lot of in-house jargon, which Greg explains here.

Why SELECT * FROM Table is Not Merely a 'Read' — Surely selecting all of the content from a table is a read operation that can be safely load balanced? Not so fast, says Hans, who comes armed with a counter-example.

PG Back Web 0.3: A Postgres Backup System with Web Interface — A Go-powered app for bringing a bit more of a friendly user interface to Postgres backups, as well as scheduled backups, backup monitoring, and webhooks. It’s available as a Docker image too.

pgcopydb 0.17: Copy a Postgres Database to a Target Postgres Server — Automates and provides a little structure around the running of pg_dump | pg_restore between two running Postgres servers.

PL/Haskell 4.0: 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 work with.