Engineering Insights

Lessons Learned Running Presto at Meta Scale

Presto is a free, open source SQL query engine. We’ve been using it at Meta for the past ten years, and learned a lot while doing so. Running anything at scale - tools, processes, services - takes problem solving to overcome unexpected challenges. Here are four things we learned while scaling up Presto to Meta scale, and some advice if you’re interested in running your own queries at scale. Scaling Presto rapidly to meet growing demands: What challenges did we face?   Deploying new Presto

Gossip Protocol Explained

You can subscribe to the system design newsletter to excel in system design interviews and software architecture. The original article was published on systemdesign.one website. What Is Gossip Protocol? The typical problems in a distributed system are the following [1], [11]: maintaining the system state (liveness of nodes) communication between nodes The potential solutions to these problems are as follows [1]: centralized state management service peer-to-pe

Fine-tuning Stable Diffusion models on Intel CPUs

Iterating from design to deploy: the shape of future builders

In a world of accelerating digital innovation, we need tools that transform the web development landscape. In his recent , Guillermo Rauch spoke about how we at Vercel enable builders—non-developers included—to tighten the cycle of design and deploy.Figma Config keynote Below, we’ll dive behind the scenes of the talk and give you tangible ways to try out Vercel’s Frontend Cloud. Read more

Performant database tree traversal with Rails

We recently solved an interesting performance problem in our Rails API when performing a tree traversal. We have some code that figures out a 3-way merge for your database schemas. To do this, we need to calculate a merge base between two schemas (like git!). Our Rails API keeps track of changes to each PlanetScale database schema. We call each of these changes a "schema snapshot," similar to a git commit that stores the state of the schema at a specific time. Each snapshot can have one or two p

Meet the Vercel AI Accelerator Participants

Today, we’re announcing the participants of —a program for the brightest builders and early-stage startups.Vercel’s AI AcceleratorAI We're thrilled to include both prominent builders and rising startups solving interesting or impactful problems, like using AI for cancer detection or transforming how academic research is made available. We received from talented startups and individuals and accepted 40, which is less than 3% of applications. The 40 accepted participants are presented below.over

Prisma 5: Faster by Default

Prisma 5.0.0 introduces new changes that make it significantly faster. These changes especially improve the experience using Prisma in serverless environments.

Accurately analyzing large scale qualitative data

Viable uses GPT-4 to analyze qualitative data at a revolutionary scale with unparalleled accuracy.