KazantsevAlexander - Fotolia

Yugabyte raises $48M to build out distributed SQL database

The leader of a young open source database company explains why the vendor is raising new money as the demand for cloud scalability continues to grow.

Open source distributed SQL database vendor Yugabyte marked a major phase of its evolution, raising $48 million in a Series B-1 round of financing led by Lightspeed Venture Partners, bringing total funding to date for the company up to $103 million.

The vendor, based in Sunnyvale, Calif., is among a group of emerging vendors in the cloud-native database market segment. Founded in 2016, Yugabyte develops an open source distributed SQL database that is compatible with the popular open source PostgreSQL database.

Last year was busy for Yugabyte, with a $30 million funding round in June and a new CEO, former Greenplum Software CEO Bill Cook, taking over the role from co-founder Kannan Muthukkaruppan.

Meanwhile, the pace of development at Yugabyte has not slowed in 2021. This year has been busy with the YugabyteDB 2.4 update on Feb. 12 bringing support for Apache Spark 3.0 to the platform.

In this Q&A, Muthukkaruppan, now president, discusses the evolution of the distributed SQL database vendor and where it is headed.

Why is Yugabyte raising new funding now and is an IPO next?

Kannan MuthukkaruppanKannan Muthukkaruppan

Kannan Muthukkaruppan: This funding is really just building on the momentum that we saw happen in 2020. Primarily, this is an indicator of the activity in the space, the market opportunity around a distributed SQL database. 

The market is looking for relational databases, like PostgreSQL, but a database that is really designed for the cloud, to give the high availability and resilience that customers want in the database tier, especially for the system of record workloads. Especially for the system of record workloads, customers want to take advantage of the horizontal scalability and elasticity of the cloud.

Our investors also take the long-term view that there's a market opportunity, and we'll continue to go down the path of building a big company here.

What has been the biggest change at Yugabyte and its distributed SQL effort over the last five years?

We're not inventing a new database API in order to keep the friction for adoption low and we get that by providing a PostgreSQL-compatible, distributed SQL database.
Kannan MuthukkaruppanPresident and co-founder, Yugabyte

Muthukkaruppan: The core vision for the company and the technology hasn't changed. We're building a database that's highly scalable, transactional and can run across multiple clouds. We're not inventing a new database API in order to keep the friction for adoption low and we get that by providing a PostgreSQL-- compatible, distributed SQL database.

We have stuck to that core belief that the database needs to be highly performant and users should not have to front the database with some type of cache, which adds more layers of complexity. One other thing where we have been pleasantly surprised that we hadn't anticipated early on is Kubernetes becoming really the platform of choice for database workloads.

Is there a need to shift the governance of the YugabyteDB open source project to an open source foundation?

Muthukkaruppan: We prefer to have the governance of the project from Yugabyte Inc., but the project nevertheless is 100% open source.

We have not kept features like encryption or backups as enterprise-only features. Everything out there is available under the Apache 2.0 open source license and we want to keep going that same way.

What do you see as the key challenges of distributed SQL?

Muthukkaruppan: A single-node architecture has some strengths, up to a certain point, while scalability is the primary weakness. But with a single-node database architecture on, say, Oracle, MySQL or PostgreSQL, there are some workloads where the performance might actually be high, because it's a much more simplified design. But what you're giving up there is the scalability.

The fixed amount of overhead in a distributed SQL database can be a little bit higher than a single node. For example, a join involving two tables might have to touch two different nodes in the cluster. So in some cases, the fixed performance of a distributed SQL database can be slightly worse than, say, the performance of single- node database. But what you get in return is the linear scalability.

Additionally with some thought process going into the transition to a distributed SQL database, it is possible to actually get really good performance.

What's next for Yugabyte?

Muthukkaruppan:  We're going to expand across all functions within the company including research and development, sales and customer functions, both in the U.S. as well as in EMEA [Europe, Middle East and Africa] and APAC [Asia Pacific]. We expect to double our headcount from its current levels of about 100 to 200-plus by the end of the year because of the strong demand we're seeing for this open source distributed SQL database.

We're also seeing demand for a fully managed offering of Yugabyte, which is the Yugabyte Cloud, and we're planning a launch for later this year. That will be a big area of investment for the company.

This interview has been edited for clarity and conciseness.

Dig Deeper on Database management

Business Analytics
Content Management