Redpanda Data Inc., creator of a streaming data platform aimed at the software developers, today announced it has raised $50 million in a new funding round, bringing its total funding to $76 million.

The three-year-old company’s product is a drop-in replacement for Apache Kafka that unifies historical and real-time data. It enables inline transformations to Amazon Web Services Inc.’s Lambda serverless platform using Kafka application program interfaces.

The software is offered as a fully managed cloud service with pricing based on throughput. “We are the only streaming vendor that does bring-your-own cloud,” said Alex Gallego, Redpanda’s founder and chief executive. “The data lives inside customer network, but we have the processing engine.”

Streaming or real-time data is data that arrives continually from sources such as sensors, process monitors and social media streams. Performing operations on real-time data requires special programs and high-speed CPUs and storage.

Built for modern hardware

The inspiration for the product was the realization that the most widely used streaming engines were designed for computers that were built years ago, Gallego said. “A fundamental change is that hard drives are 1,000 times faster today and computers have more CPU cores,” he said. “The bottleneck has shifted from spinning disk to CPU coordination.”

Redpanda says it provides a single, secure and limitlessly scalable platform for processing both historical and real-time data thanks to a new approach to tiered storage and onboard data transformation. Traditional stream processing has often involved sacrificing performance to ensure data was not lost. With Redpanda, Gallego said, “you can have both.”

Writing software to process streaming data has traditionally involved manually cobbling together numerous discrete components. “We remove all external dependencies; it’s a single binary,” Gallego said. “We can be 10 times faster. I call us layer 0; every single event that happens in your network uses Redpanda as the bottom layer.”

The company’s immediate opportunity is to replace legacy platforms such as Kafka, Apache Pulsar and RabbitMQ, while the bigger goal is “to fundamentally reshape and extend how organizations make use of streaming data,” he said.

Redpanda is available as a self-hosted binary under a “source available” license that gives the company four-year exclusivity to offer the product as a service. After four years, each release reverts to a standard Apache license.

Word-of-mouth marketing

Awareness has spread rapidly by word of mouth, Gallego said. The software was originally posted on GitHub in November 2020. “By end of the year there were several thousand clusters” in operation, he said. “By the end of last year we were at hundreds of thousands of clusters deployed and today we do as many clusters each day as we did in our first six months.”

The company, which has about 60 employees, expects to spend much of the new investment to build out the platform as well as fortify sales and marketing. “We want to be a leader in a bottom-up movement led by developers,” Gallego said. Its long-term plans are to deliver an API that ties directly into serverless deployment.

“Developers don’t care what the technology is; they just want to get the job done,” Gallego said. “We give them both compatibility with 10 years of Kafka and the future of streaming.”

The Series B funding was led by GV Management Co LLC, formerly Google Ventures, with participation from Lightspeed Venture Partners and Haystack Management Co. LLC.

Image: Redpanda