New York City-based startup Mortar Data has closed a $1.8 million seed round for its service that it claims can have even Hadoop newbies writing jobs in an hour. Genacast Ventures led the round, which also included Atlas Ventures, Great Oaks Ventures, and a slew of individual investors. Mortar Data emerged from stealth mode in April 2012, and is still available on an invite-only basis.
The idea behind the company is that developers and data scientists without huge budgets and/or mad Java skills should be able to take advantage of Hadoop both cheaply and easily. Mortar Data’s trick is a custom blend of the Pig and Python programming languages that makes job creation a snap, coupled with a cloud computing backend that means users don’t even have to own a server.
However, because it’s hosted on the Amazon Web Services cloud, Mortar Data is still best-suited to customers already storing data in Amazon’s S3 storage service. The good news for the company at this early stage is there are plenty of developers doing just that. It’s also good news for the developer community that Mortar recognizes this limitation and is working to expand the number of data sources and repositories it supports, as well as the means of visualizing the results of users’ Hadoop jobs.
In the meantime, you can see some of what Mortar is up to on its Github page, which includes the development framework (so developers can write jobs locally), a MongoDB adapter, an API client for Ruby and a couple of sample datasets. And, Co-Founder and CEO K Young has promised me that we can expect to see more news out of Mortar Data later this month as the company begins ramping up its efforts publicly.
Here’s a video of demonstrating the service in action:
[vimeo http://www.vimeo.com/51017287 w=400&h=300]