Engineering Insights

The Slotted Counter Pattern

It is a common database pattern to increment an INT column when an event happens, such as a download or page view. You can go far with this pattern until bursts of these types of events happen in parallel and you experience contention on a single row. When multiple transactions are trying to update the counter, you are essentially forcing these transactions to run serially, which is bad for concurrency and can cause deadlocks. You can also see dramatic increases in query time when bursts like th

Build your own web framework

Have you ever wondered what it takes to build your own web framework that also deploys to edge and serverless infrastructure? What features does a modern framework need to support, and how can we ensure that these features allow us to build a scalable, performant web application? Read more

Efficient training of language models to fill in the middle

Introducing new audio and vision documentation in 🤗 Datasets

Behind the scenes: How we built Password Roles

We recently released a new feature that allows you to use granular roles for your database passwords. When you generate a new password, you have the option to select from the following roles: read-only, write-only, read/write, admin. We implemented password roles using Vitess Access Control Lists and VTTablet, but we ran into a couple of roadblocks on the way. In this post, we look at some of the issues we faced while implementing this, and how we were able to get around them. How Vitess handles

Faster Text Generation with TensorFlow and XLA

Node.js 18.7.0 (Current)

Safely dropping MySQL tables

Dropping or removing an unused table from a database schema can be challenging. Even after triple checking that all of your apps have migrated away from querying the table, there may still be that one rogue script that accesses it. Running the dreaded DROP TABLE statement can cause a host of unintentional problems if that table is still being used elsewhere. It completely erases the table definition, partitions, and the data in that table. Before you drop a table, you should double (or triple!)