BlazingDB, the distributed GPU-accelerated data warehouse, today announced a $2.9 million seed round with participation from NVIDIA GPU Ventures and Samsung NEXT. NVIDIA and Samsung NEXT bring a lot more than funding to the table, with strategic support in technical, enterprise sales, and marketing expertise. To support BlazingDB’s growth into enterprise, NVIDIA will also be granted a seat on BlazingDB’s board of directors. Samsung NEXT will have a board observer seat and will act as a strategic advisor to the company.
“With this round of financing, we’ll expand our team and accelerate enterprise deployments,” says Rodrigo Aramburu, CEO at BlazingDB. “Investors like NVIDIA and Samsung NEXT grant us exposure to enterprise and technology expertise, allowing us to grow faster. Additionally, Samsung NEXT will provide office space alongside fellow portfolio companies, where we can leverage the knowledge of other founders and a team of serial entrepreneurs.”
BlazingDB has built a state-of-the-art GPU-accelerated SQL analytics engine able to process massive datasets stored on common data lake technologies already implemented within enterprise organizations. BlazingDB is a ground up redesign of relational database architecture, and has been able to outperform some of the most performant SQL data warehouse offerings in enterprise by over 10x.
“BlazingDB is pioneering the field of accelerated analytics across massive distributed data lakes, which will solve major pain points for the industry,” said Jeff Herbst, vice president of business development at NVIDIA. “Its unique GPU-powered SQL technology is highly complementary to our other ecosystem partners. We’re thrilled to expand our partnership with them through this investment.”
“The power of BlazingDB lies in its ability to leverage GPUs to make querying enormous data stores super-fast and easy-to-use,” said Leo Chang, Principal at Samsung NEXT and lead investor. “At Samsung NEXT, we understand the importance of transformative software and services; the technical ambition of BlazingDB demonstrates the magnitude and challenge of bringing value from large amounts of data, quickly.”
With this round of financing BlazingDB will add new team members to continue to build out the core engine, automated cloud deployment tooling, and more for version 2.0. BlazingDB 2.0 enables customers to run SQL directly on Apache Parquet datasets already inside data lakes. Customers will no longer need to import hundreds of terabytes of data into BlazingDB, they’ll be able to run queries off of live massive data in a matter of minutes.
This money will also enable BlazingDB to ramp up sales activity with enterprise users that are currently leveraging BlazingDB to operationalize tens to hundreds of terabytes of data inside their data lakes with SQL analytics.
BlazingDB was founded in June 2015 by the Aramburu brothers, Felipe (CTO) and Rodrigo (CEO). The project started through their consulting firm in Lima, Peru, where they were performing multi-terabyte SQL joins, and nothing could performantly run their queries. They built a GPU table joiner and ran 35-hour queries in 30 seconds, a 4,000x performance improvement. Since then BlazingDB has gone through the Techstars accelerator program in Boulder, were TechCrunch Disrupt SF Battlefield finalists in 2016, and have been working with multiple enterprise data engineering teams building out their next generation of data pipelines.