October Product Update

2 min read

•

2 days ago

As traffic spikes approach for the holiday season, we’ve rolled out new improvements to help you keep database performance fast and predictable. Here’s what’s new:

💬 Product Updates

Readyset Core

  • Smarter Time-Series Aggregations: We introduced Bucket(timestamptz, interval), a Readyset-specific SQL function that groups timestamps into fixed intervals. It’s ideal for time-series queries like metrics rollups or activity heatmaps. This makes it easier to analyze trends over time without restructuring your queries or data.
  • String Aggregates: We added full support for PostgreSQL’s string_agg() function, enabling users to efficiently concatenate string values in grouped queries. This is particularly useful for generating comma-separated lists or compact summaries within SQL.
  • Better Accuracy with NULLs and DISTINCT: Accumulating aggregation functions (like group_concat() and array_agg()) now ignore NULL values and support the DISTINCT keyword, ensuring correctness and flexibility when summarizing grouped results.
  • ORDER BY: ORDER BY clauses inside aggregation functions are now correctly handled, allowing developers to control the order of concatenated or accumulated values. This is crucial for use cases like ordered logs, ranked lists, or other sequence-sensitive use cases.
  • Catch Misconfigurations Early: We've added pre-snapshot verification checks to surface potential config issues before they cause runtime errors. Use --verify to validate your setup proactively or skip checks with --verify-skip for advanced workflows.
  • And more!

Check out the full release here. 

🆕 New on the Blog

Does Your Black Friday Database Scaling Strategy Involve Duct Tape and Prayers?

If your scaling plan relies on last-minute patches or extra hardware, this one’s for you. We explain why traditional strategies fall short during traffic spikes and how Readyset helps you survive Black Friday without the stress.

How Readyset Handles Window Functions Under the Hood

Window functions are powerful but tricky to support in a high-performance cache. This deep dive explains how we built support for ROW_NUMBER, RANK, and more while preserving Readyset’s speed and correctness.

October Event Recap

MySQL Brazil Conf 

We were thrilled to be part of this year’s MySQL Brazil Conference, where Marcelo co-organized the event and Tanmay and Bianchi presented the opening keynote Beyond Read Replicas: Introducing Readyset QueryPilot for MySQL Scaling. 

Huge thanks to the organizers, speakers, and everyone who joined our session! 

🗓️ Upcoming Events

Webinar with Tanmay and Bianchi - November 6, 2025

Big sale coming up? Don’t let your database be the reason customers can’t check out.

With Black Friday around the corner, join Tanmay and Bianchi to see how QueryPilot keeps performance predictable and databases stable without replicas, rewrites, or sleepless nights.

Register here for the webinar next Thursday 11/6. 

🧑‍💻Request Features

Keen to use Readyset but your database or cloud environment are not supported? Reach out to support@readyset.io and let us know.

🤝 Get Involved

We love hearing from you! If you want to get involved with Readyset, you can:

Authors

R

Readyset