April Product Update

3 min read

•

19 days ago

Hi there, hope your spring is off to a great start! Here’s what’s new at Readyset:

🎉 Readyset Private on Azure VM (MySQL Edition)

You can now deploy Readyset privately on an Azure VM with MySQL and see performance gains from milliseconds to microseconds! Setting up takes just minutes and a few commands, with no replicas, rewrites, or added complexity required.

It’s a perfect fit for teams running on Azure and seeking to enhance MySQL performance and scalability while reducing costs. All of this is achieved without the burden of managing replicas, rewrites, or additional infrastructure complexity.

Read the guide here for all the details. 

🚀 Product Updates 

Readyset Cloud

  • Deployed an LLM-based log analysis utility for diagnosing cluster creation failures, substantially reducing issue triage time.
  • Refactored the cluster creation and update forms, improving maintainability and user experience.
  • Migrated Readyset Cloud from EKS to ECS, reducing infrastructure complexity.
  • Added automated TTL enforcement for workflow and deleted cluster resources to improve lifecycle hygiene.
  • Upgraded the frontend to the latest version of Next.js for improved stability and performance.
  • And more! 

Readyset Core

  • As a first step toward supporting latin1 text columns in MySQL, we now allow replicating these columns but storing them re-encoded as UTF-8. Data will be retrieved as UTF-8, so comparisons and retrieval will not match upstream. Future work will improve support.
  • Resolved a MySQL issue that occurred when Minimal Row-Based Replication was enabled and the user ran an INSERT with no columns, thereby improving stability and correctness for common write patterns under newer MySQL configurations.
  • Fixed a MySQL issue where INSERT statements with omitted columns could cause replication to fail under certain conditions, ensuring better reliability and broader support for partially-specified inserts.
  • Corrected an issue with MySQL Minimal Row Based Replication where the collation and signedness of columns were incorrectly retrieved from the table map event, preventing type mismatches.
  • Fixed a bug where specific unsupported queries would throw an error instead of getting proxied upstream, improving user experience and ensuring seamless fallback behavior.
  • And more! 

Check out the full release here. 

🆕 New on the Blog 

  • Vertical Scaling of MySQL vs. Horizontal Scaling with Readyset: What’s the smartest way to scale MySQL in the cloud? In this post, Vinicius Grippa analyzes real-world benchmarks and cost data, comparing vertically scaling with larger instances to horizontally scaling using Readyset. The results? Readyset delivered 155% better throughput with over 50% lower cost per QPS.
  • A Deep Dive on Readyset Behind PgBouncer: Bianchi shares a hands-on guide to deploying Readyset behind PgBouncer to optimize PostgreSQL performance. Routing reads to Readyset and writes to the primary database boosts performance and reduces infrastructure costs. The post includes real-world config examples, a sample workload, and benchmarks to help you get started.
  • Tired of slow queries and expensive replicas? Readyset solves both, and we’ve got the receipts (well, the YouTube shorts). Readyset Caching | Readyset’s Massive Cost Savings 

🐬 MySQL HeatWave Summit Recap

Tanmay, Marcelo, and I had a fantastic time celebrating MySQL’s 30th anniversary and connecting with the community at HeatWave Summit!

From presenting “Turbocharging Your MySQL with Readyset” to diving into insightful conversations on database performance and sharing plenty of laughs at the community dinner, it was an incredible two days.

We’ve already marked our calendars for the next one!

🗓️ Upcoming Events

PGConf Germany: May 8-9

Tanmay and I are excited to be attending PGConf Germany in Berlin, and we’re looking forward to connecting on all things Postgres performance! Come find us at the Cybertec booth, we’d love to meet you.

PGDay Blumenau: May 24

We’re proud to be sponsoring PGDay Blumenau, where Marcelo and Bianchi will be presenting “Scaling Reads in PostgreSQL with Readyset: High-Performance Intelligent Caching Without Invalidation.”

If you’re attending, stop by for great technical insights on scaling PostgreSQL efficiently!

🧑‍💻Request Features

Keen to use Readyset but your database or cloud environment is 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