---
title: "Latency's huge impact on Postgres performance"
publication: "Postgres Weekly"
issue: 630
published_at: "2026-01-07"
canonical_url: "https://postgresweekly.com/issues/630"
---

Plus a review of the top ten items of 2025. |        

#​630 — January 7, 2026 

[Web Version](https://postgresweekly.com/issues/630)   

🎉 Happy New Year! A quick reminder that Postgres Weekly is now sent every Wednesday — starting with this very issue. 
__
_Your editor, Peter Cooper_ 

Together with [![Microsoft logo\](https://res.cloudinary.com/cpress/image/upload/c_fill,g_auto,w_280,h_60/e_make_transparent/co_white,e_outline:7/om9hjsmudy5lsvba59eb.png)](https://posetteconf.com/2026/cfp/)   

Postgres Weekly 

[![\](https://res.cloudinary.com/cpress/image/upload/w_1280,e_sharpen:60,q_auto/hsvy3genjwv9bfv9klsi.jpg)](https://www.cs.cmu.edu/~pavlo/blog/2026/01/2025-databases-retrospective.html)  

[Databases in 2025: A Year in Review](https://www.cs.cmu.edu/~pavlo/blog/2026/01/2025-databases-retrospective.html) — Esteemed database expert Andy Pavlo has put together a fantastic look at the major trends and things that went on in the world of databases over the past year. Postgres gets pride of place as the ‘dominant’ database of 2025, but there’s much more to enjoy here. 

Andy Pavlo   [![\](https://res.cloudinary.com/cpress/image/upload/c_limit,w_480,h_480,q_auto/copm/a0362e9b.png)](https://posetteconf.com/2026/cfp/) 

[POSETTE 2026—The Call for Proposals Closes Soon!](https://posetteconf.com/2026/cfp/) — New year, new goals! Resolve to share your Postgres insights at [POSETTE: An Event for Postgres](https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fposetteconf.com%2F2026%2F&data=05|02|Teresa.Giacomini%40microsoft.com|3d64356a1f584781bd3808de3f4d5984|72f988bf86f141af91ab2d7cd011db47|1|0|639017798611639640|Unknown|TWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D|0|||&sdata=jWRYW8hP7Cnh%2BKfnXirwgwri%2FupvebbIpfhn%2BtLEkwE%3D&reserved=0), a free & virtual developer event organized by the PostgreSQL team at Microsoft—happening June 16–18, 2026. The **CFP closes on Feb 1**. [Get started](https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fposetteconf.com%2F2026%2Fcfp%2F&data=05|02|Teresa.Giacomini%40microsoft.com|3d64356a1f584781bd3808de3f4d5984|72f988bf86f141af91ab2d7cd011db47|1|0|639017798611659539|Unknown|TWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D|0|||&sdata=9ktS9ps0Ho8HnHQTf%2BB7mgkrHQjKyi9QKevEFYYHYVA%3D&reserved=0). 

Microsoft sponsor   

**IN BRIEF:** 

- 🇨🇭 [CERN PGDay 2026](https://www.swisspug.org/cern-pgday-2026.html) takes place in Switzerland this February 6 and the [full schedule](https://indico.cern.ch/event/1504097/timetable/#20260206) has been published.
- [The Postgres team has recognized](https://www.postgresql.org/about/news/new-postgresql-contributors-december-2025-3200/) 12 new people as _significant contributors_ to the project, as well as announcing Floor Drees has joined the contributors team _"to help with spreading the workload."_   

[Postgres Performance: Latency in the Cloud & On-Premise](https://www.cybertec-postgresql.com/en/postgresql-performance-latency-in-the-cloud-and-on-premise/) — Having a performant database is one thing, but its _location_ has a big impact on latency — which, if too high, can outweigh everything else performance-wise. 

Hans-Jürgen Schönig   

[Postgres 18's `RETURNING` Enhancements](https://www.pgedge.com/blog/postgresql-18-returning-enhancements-a-game-changer-for-modern-applications) — Postgres 18 enhanced the `RETURNING` clause used in DML queries (inserting, updating, etc.) to support `old` and `new` aliases to explicitly access the previous and current state of data. 

Ahsan Hadi   

[pg_csv_loader: A Tool for 'Quick and Dirty' Loading of CSV Files](https://www.depesz.com/2026/01/07/quick-and-dirty-loading-of-csv-files/) — A JavaScript-powered tool for loading CSV files into Postgres databases with minimal configuration needed. [Source.](https://gitlab.com/depesz/pg_csv_loader) 

Hubert depesz Lubaczewski   

📄 [Instant Database Clones with Postgres 18](https://boringsql.com/posts/instant-database-clones/) – By way of using [template databases.](https://www.postgresql.org/docs/current/manage-ag-templatedbs.html) Radim Marek 

📄 [A Few Quick Postgres Scripting Tips](https://www.pgrs.net/2026/01/06/postgresql-scripting-tips/) Paul Gross 

📄 [The State of Postgres MCP Servers in 2025](https://dbhub.ai/blog/state-of-postgres-mcp-servers-2025) DBHub      

📰 Classifieds   

🚀 [Ship AI agents to production](https://www.pgedge.com/products/agentic-ai-postgres). Postgres toolkit with MCP, hybrid search & global scale. [Beta now open](https://www.pgedge.com/products/agentic-ai-postgres).      

🐘 **THE TOP POSTGRES WEEKLY ITEMS OF 2025:** 

In the [final issue of 2025](https://postgresweekly.com/issues/629) we took a look back at the top tools and code items from the past year, but here's the broader view of what got the most engagement during the past twelve months. It's a list of gems well worth revisiting:   

1. [Life-Altering Postgres Patterns](https://mccue.dev/pages/3-11-25-life-altering-postgresql-patterns) — The author promised us that the title isn’t just clickbait (although clearly it worked, being at #1!) and delivered twelve different bite-size, hard-learned tips and insights in areas from using UUIDs as primary keys and table naming to the use of schemas and views. 

Ethan McCue   

2. [Don't Do This (in Postgres)](https://wiki.postgresql.org/wiki/Don%27t_Do_This) — A perennially interesting page on the official Postgres wiki rounding up advice around ‘common mistakes’ in using Postgres and what _not_ to do, like _“Don’t use `char(n)`”_ and _“Don’t use `serial`“_. Some will be contentious, but are backed up with reasons. 

Postgres Wiki   

3. [Kafka is Fast, I'll Use Postgres](https://topicpartition.io/blog/postgres-pubsub-queue-benchmarks) — Inspired by [another post](https://dizzy.zone/2025/09/24/Redis-is-fast-Ill-cache-in-Postgres/) about using Postgres instead of Redis for caching, the author sets out to see if Postgres is good _enough_ to handle use cases for which you might naturally choose [Kafka.](https://kafka.apache.org/) 

Stanislav Kozlovski   

4. [How to Fix a Common Cause of Slow Queries in Postgres](https://render.com/blog/postgresql-top-cause-slow-queries) — A database engineer at _Render_ shows off a commonly encountered, but trivially fixed, performance issue caused by missing indexes on foreign keys. 

Eric Fritz   

5. [Lessons from Scaling Postgres Queues to 100K Events Per Second](https://www.rudderstack.com/blog/scaling-postgres-queue/) — RudderStack decided to use Postgres as their main queueing system rather than something like Kafka and their team shares the story of what they learned. 

Aris Tzoumas (RudderStack)   

6. [Benchmarking Postgres 17 vs 18](https://planetscale.com/blog/benchmarking-postgres-17-vs-18) — The author conducted a series of detailed performance benchmarks between Postgres 17 and 18, in some 96 combinations, and reassuringly finds that Postgres 18 provides a nice performance bump, that local disks rule, and tweaking your settings remains worthwhile. 

Ben Dicken (PlanetScale)   

7. [1 Trillion Rows in Citus?](https://www.cybertec-postgresql.com/en/postgresql-1-trillion-rows-in-citus/) — _“Postgres scales,”_ but just how far does that go? Hans-Jürgen decided to find out, running a little (or is it big?) experiment to see if a 1 trillion row table is even achievable. 

Hans-Jürgen Schönig   

8. [Just Because You’re Getting an Index Scan, Doesn't Mean You Can’t Do Better](https://www.pgmustard.com/blog/index-scan-doesnt-mean-its-fast) — If you see index scans when reviewing query plans, you might think you’re on the right path to a high performing query, but there’s more juice left for the squeezing, says Michael. 

Michael Christofides   

9. [Pipelining Comes to `psql` in Postgres 18](https://postgresql.verite.pro/blog/2025/10/01/psql-pipeline.html) — `psql` in Postgres 18 and later is equipped with commands to use pipelining in SQL scripts, something Daniel notes _“can provide huge gains in query throughput”._ 

Daniel Vérité   

10. [Postgres in the Time of Monster Hardware](https://www.enterprisedb.com/blog/postgres-time-monster-hardware) — You might think your workstation's CPU is pretty fast, but imagine having an AMD EPYC with 192 cores per socket and 10 terabytes of RAM! Modern CPU power (not to mention much faster storage) invites us to ask questions about modern ways to scale database servers. 

Lætitia Avrot    

**Got a new email?** You can [change your address here](https://postgresweekly.com/edit_subscription/*|UID|*). 

--- 

**Bored of us?** [Click here to stop getting Postgres Weekly](https://postgresweekly.com/leave/*|UID|*). 

--- 

**Got a link for us?** Reply and tell us. We can't include everything but we'll look at anything you send. _Thanks!_ 

--- 

Published by Cooper Press Ltd.
Fairfield Enterprise Centre, Louth, LN11 0LS, United Kingdom 

Postgres, PostgreSQL and the Slonik Logo are trademarks or registered trademarks of the PostgreSQL Community Association of Canada. 

[Stop getting this newsletter](https://postgresweekly.com/leave/*|UID|*).
