Quest, the big data analytics platform from Metanautix, erases the barriers between data types by representing any kind of information as relational tables compatible with SQL queries.
Making SQL, NoSQL, Hadoop and other big data frameworks play nicely with one another is a major challenge that vendors are only now beginning to overcome. But a startup named Metanautix is taking data-agnosticism even further through a new platform that can turn any kind of data—even images—into SQL tables.
The company, which launched last August with $7 million in Series A financing, bills its platform, called Quest, as "a whole new way to intuitively navigate and analyze any data, from any source, at any scale."
That's a weighty promise for an ecosystem where the nature of data and the way it's stored often require different data analysis tools and strategies. But by allowing all types of data to be processed using SQL queries, which are among the most familiar tools for data analysis, Quest supports data analytics on any type of information without requiring expertise in special analytics software or languages, according to Metanautix.
The platform's data-agnosticism extends even to visual data, as the company explained in a recent blog post. Quest can map the pixels of images in a variety of formats into relational tables, making them compatible with SQL queries.
Of course, working with images pixel by pixel may not always allow data scientists to do what they need with visual data. And it's not clear that the strategy could be feasibly extended to video files, which would require mapping a magnitudally larger number of pixels.
Still, Metanautix is doing interesting things by pushing data agnosticism to the next level. Moving forward, the big data conversation likely will become not about how to make different data analytics and storage tools compatible with each other, but how to make analytics strategies work with any type of data.