PipelineDB, a Y Combinator Winter 2014 graduate, announced the availability of the open source version of its streaming SQL database product today. A commercial version is expected later this year.
The product is an open-source database that runs SQL queries continuously in streams, incrementally storing the results in tables, company co-founder Derek Nelson explained. “[The product] unifies continuous processing and relational storage, making it unnecessary to manage an external storage system alongside the stream-processing system,” he added.
The typical use case for this product involves reporting and analytics and real-time monitoring, he said. For instance, an e-commerce website running A/B testing might use the product to get the results of a change in real time, as opposed to waiting until the next day.
Nelson claims that this product can do away with ETL (extract, transform and load), which has been the way databases have been traditionally created. PipelineDB wants to make it a more streamlined process that happens in real time.
The company differentiates itself from the database pack by offering a streaming product that uses SQL queries instead of a special programming language. That means anyone familiar with SQL queries should be able to get up and running with the product quickly.
Nelson came up with the idea for the company when he was working at his former job at AdRoll and kept wishing he had a tool with this functionality. He finally decided to leave the company and create the tool himself and PipelineDB was born.
He said the four-employee company made the decision to offer an open source version of the product for rather pragmatic reasons. First of all, they realized that it’s difficult to get a company to place a long-term bet on a startup and offering an open source version gives a potential customer time to get comfortable with the product.
Secondly, Nelson said that the biggest technology companies like Facebook, Google and Amazon are obsessed with open source and he can’t expect to compete with that with a proprietary product.
“Being a part of Y Combinator was awesome,” Nelson said. The company was probably the most technical in the class, which made it a bit of a black sheep, but also made it easier for them because they weren’t competing with more commercial offerings.
Overall he says, being part of YC helped them build a network, get into companies for pre-sales conversations and it gave them access to the wisdom and experience of the YC partners, which was also extremely valuable, Nelson said.
Today, with the release of the open source product, the company took another step on that journey.
So far, PipelineDB has received a seed funding round. It chose not to disclose the amount, but investors included SV Angel, Data Collective, Paul Buchheit, Susa Ventures, TenOneTen, and a few other angels Nelson didn’t name.