~postgresweekly | Issue 565 - The fastest way to copy data between tables (8)

pg_statviz: Time Series Analysis and Visualization of Postgres Stats — An extension and utility pair for performing time series analysis and visualization of various stats around things like I/O, locks, WAL, and buffer use. The new v0.7 release adds Postgres 17 support, new I/O stats, and Pandas resampling for collected data.

The Fastest Way to Copy Data Between Postgres Tables — Anthony compares several methods, both using native SQL and extensions. The spoiler is that pg_bulkload wins by quite a way, though with some tradeoffs.

Find the Root Cause of Postgres Performance Problems — Slow queries making your application feel sluggish? Bloated tables causing unexpected I/O? Use pganalyze to deliver consistent database performance & availability through intelligent tuning advisors and in-depth monitoring.

ClickHouse Acquires PeerDB for Native Postgres CDC Integration — PeerDB is a data replication tool built by former Microsoft and Citus Data employees for streaming data from Postgres to data warehouses, queues, and other storage engines. It seems data warehouse company ClickHouse has seen the strategic value in bringing such tech in-house.

Designing a Prototype: Postgres Plan Freezing — Plan freezing is the idea of saving a specific plan for a parameterized query so it can be reused again in future as-is. Andrei explains the technicalities of implementing his extension (which you can also try out) to do this.

What to Know About Linux Memory Overcommit with Postgres — Linux’s overcommit feature can allow for more memory to be allocated than is actually available. This has benefits in some cases, but it’s bad news for Postgres.

How to Keep a Postgres Community Alive — Fresh off the excitement of PG Day France, the author shares his thoughts on building up a local Postgres user group and running its meetings.

Graph Retrieval with Recursive CTEs — Postgres isn’t a ‘graph database’ but it can certainly handle graph related concepts.