How Twitter Runs Presto at Scale in the Cloud

Beinan Wang, Ph.D. - Twitter - Presto Community Track - Percona Live 2021

Presto is a widely adopted federated SQL engine for federated querying across multiple data sources. With Presto, you can perform ad hoc querying of data in place.

In this session, Twitter engineer Beinan Wang will share how they use Presto at scale with over 3K Presto workers and 10 million queries and a highly-scalable query predictor service. At Twitter, this service helps improve the performance of Presto clusters and provides expected execution statistics on Business Intelligence dashboards. ∎

Video

Link: https://youtu.be/CoNgDcyl-IE

Speakers

Beinan Wang, Ph.D.

Twitter, Sr. Software Engineer

Beinan builds large scale distributed SQL systems (presto&hive) for Twitter’s data platform team.

See all talks by Beinan Wang, Ph.D. »

Did you like this post? Why not read more?

Percona Community Live June 21 - Day 1

You are invited to participate in the first day of the Percona Community Live stream! Learn about VictoriaMetrics, flexible Indexing with Postgres, PostgreSQL backups, database resiliency, and many other things on June 21st. Join us on June 22nd and June 23d for more amazing content!

Data Gravity Episode 3 - benny Vasquez - AlmaLinux Project and the Future of Databases

In this episode, Joe Brockmeier interviews benny Vasquez of the AlmaLinux Project to discuss open source data, the future of databases, and how databases and Linux distros go together.

Data Gravity Episode 2 - Sonia Valeja and David Gonzalez - PostgreSQL for Jobseekers Book

In this episode, Joe Brockmeier talks to Percona’s Sonia Valeja and David Gonzalez, about their recent book PostgreSQL for Jobseekers. They talk about the inspiration for the book, why they focus on jobseekers, and what’s covered in the title.