Fueling Presto’s Momentum and IBM’s Growing Role in the Open-Source SQL Engine

    By: 

    Ethan Zhang, Program Director, watsonx.data at IBM

    I’ve now been a part of IBM for 2 years and I’m pretty encouraged with the work this team has put into open-source Presto. So, I wanted to take some time to share in a blog what we’ve been up to for the last 2 years and the growth we’ve seen collectively in the community and project. 

    Since joining the Presto community, IBM has steadily increased its contributions both in code and in collaboration. What started as a few engineers experimenting with the engine has grown into a major engineering investment, community leadership role, and a clear vision for accelerating Presto’s next chapter. 

    A Clear Upward Trend 

    IBM’s contributions to Presto have grown steadily over the past two years. Since acquiring Ahana in July 2023 (where I came from), monthly IBM-authored pull requests (PRs) have increased threefold. In that same timeframe, IBM’s PR merge count rose by 440%, reflecting a major spike in direct contributions. 

    The project is seeing increased activity as well: total PRs merged in Presto grew by 39%, and the average time to close PRs decreased by 6.57%, suggesting a more efficient and engaged community. On IBM’s side, PR closure time improved by nearly 34%, meaning IBM’s contributions are moving faster through review and into production. 

    And this growth reflects not just quantity but increasing technical depth. IBM is now leading several major efforts in Presto, including: 

    Velox + Presto: Building Presto C++ 

    Much of this innovation is powered by Velox, Meta’s open-source vectorized execution engine, which IBM actively contributes to. Presto C++ is built on Velox and offers a dramatic step forward in query performance and efficiency. 

    Just last week we were at VeloxCon 2025, an incredible event with close to 500 people there (and over 1.5K livestream views!) to showcase some of our work including Presto C++ features, operational efficiency, and Iceberg support. 

    The Reality of Presto’s Momentum 

    Despite claims to the contrary, Presto is not a legacy project on its way out. I’d in fact argue the opposite – Presto is thriving. 

    Today, Presto has an active, growing contributor base led by engineers at IBM, Meta, Uber, Denodo, and more, with an expanding set of production users. Over the last few months alone, organizations like Twilio, Zain Sudan, Jio Platforms, and Bolt have publicly shared how they’re using Presto in their data infra. 

    What’s Next 

    Make sure you register for PrestoCon Day on June 17, our virtual community event where we’ll be covering a lot of this work! There will also be many other companies there sharing their Presto work.

    IBM is all-in on open-source data infrastructure. Presto is just one of our key data infra engines we’re focused on (we’re also doing a ton in Velox and Apache Gluten). We’re build the core layers of the open data lakehouse in the open, with the community. Join us!