Upsolver raises $25M for no-code data lake platform

The market and demand for technologies that help organizations make effective use of cloud data lakes are growing fast. Among the fast-growing vendors in this arena is Upsolver, which revealed on Tuesday that it raised $25 million in a Series B round of funding.

The funding round was led by Scale Venture Partners and included the participation of JVP, Vertex Ventures US and Wing Venture Capital. The new funding comes less than a year after Upsolver's $13 million Series A in June 2020.

Upsolver's cloud data lake-enabling technology provides users with a no-code platform to transform data so that it can be queried to help make decisions. The promise of no-code is a drag-and-drop user interface that enables complex configurations without users needing to program interactions.

In this Q&A, Ori Rafael, CEO and co-founder of Upsolver, based in Sunnyvale, Calif., discusses the cloud data lake marketplace and what data engineering is all about.

Why are you now raising a Series B for Upsolver?

Ori Rafael: We didn't plan on raising again so soon after the Series A round that only closed in June 2020. But it was a very strong year growth-wise, so we decided to raise much earlier than we originally planned.

Ori Rafael

We are basically expanding on all on all fronts. We're going to hire more and we're going to bring in the people that we need for scale. I think it's about doing things faster; that was the reason for doing the round and to build a stronger team.

How has the business of Upsolver changed since you helped start the company in 2014?

Rafael: When we founded Upsolver, we started in the advertising business and ended up building our own database for advertising purposes. Three years ago we pivoted the company using the product we built internally to iterate faster with data lakes.

I think from the market perspective, it has been really amazing to see how data lakes have been so widely adopted in the last few years. When we got started, we had to explain what a data lake is to people, and now everyone has a data lake.

It's really amazing to see how much the data lake area has grown, from three years ago to now, it's a completely different story. It's really the golden age of data.

It's really amazing to see how much the data lake area has grown, from three years ago to now, it's a completely different story. It's really the golden age of data.

What is a data lake engineering platform? Is it a function of data middleware?

Rafael: With the word middleware, people tend imagine something very traditional from the enterprise world. The term data middleware isn't specific enough for us.

We are in the domain of transforming raw data into data that people can actually use. It's kind of a hybrid between the old ETL [extract, transform, load] world and the database world. 

We call Upsolver a no-code data lake engineering platform. The closest comparison would be open source platforms like Hadoop or Spark. They're also helping you to manage the data lake, but what they're doing is a very code-intensive process, with hundreds of configurations you need to go through, such that you need to have very skilled big data engineers to use those technologies. Upsolver is a no-code alternative. So we are providing the benefits of using a data lake in a no-code approach.

What role do you see for Presto for the data lake engineering platform at Upsolver?

Rafael: Presto is an engine used to query a data lake and Upsolver is an engine used to build a data lake. Together it's a solution that provides the same value as a database.

We work closely with different Presto distributions, including Amazon Athena, which is the AWS service for managed Presto. Presto is a very good use case for us and that's why we also decided to join the Presto community and I'm on the Governing Board of the Presto Foundation.

Ahana, just like Amazon Athena, is a Presto query engine that queries the data lake. So any query engine that queries the data lake is a natural partner for us.

We are already seeing that there are a whole bunch of query engines on top of the data lake. So there is Trino and Presto, and commercially there is Ahana, Starburst, Dremio and Amazon Athena -- so it makes sense that we are going to have many different query engines in the market.