Yahoo has been widely credited with helping create the Big Data phenomenon by popularizing the open source Hadoop data processing platform. So it's no wonder that dozens of Yahoo alumni have gone on to found companies that are considered stars of the Big Data universe, including Cloudera, which earlier this year raised $160 million to bring its fundraising total to $300 million.
The executive team of startup BitYota, which delivers a cloud-based data warehouse-as-a-service, is packed with Yahoo alums, including CEO Dev Patel, CTO Harmeek Singh Bedi, VP of Product Poulomi Damany and VP of Engineering Paresh Goswami.
The Yahoo pedigree is evident at BitYota, Damany said. The Mountain View, Calif., company's 20 employees are "focused on doing groundbreaking new stuff" while always mindful of how to scale a company for steady growth.
Not the Same Old Data Warehouse
BitYota's product is fundamentally different from better-known cloud data warehouse services such as Amazon's Redshift, Damany said. Redshift is built on massively parallel processing (MPP) technology from ParAccel. While this allows Amazon to utilize inexpensive commodity hardware instead of pricey appliances and offer a low-cost service, Damany said it does not really account for the increasingly diverse types of data that companies want to analyze.
The Redshift model "relies on structured data," she said. "It assumes you have a high performance data node with a homogeneous structure." In contrast, BitYota promises to allow companies to query unstructured data with no need for ETL (extract, transform, load) processes or upfront data modeling.
Damany said BitYota serves two types of customers: "born in the cloud" companies that have built cloud infrastructures and rely heavily on Web and/or mobile apps, and companies with traditional on-premises infrastructures that are moving some applications to the cloud and gathering growing amounts of customer data through Web and mobile channels.
The first group is comfortable with storing data in the cloud and familiar with semi-structured data but less familiar with data modeling and data architecture concepts. BitYota's technology "allows them to discover attributes they care about and potentially create dashboards out of it," Damany said. The second group is more interested in combining new data types with their existing data sources such as ERP systems, then deriving insights from the combined data.
A key component of BitYota's technology helps both types of companies by employing a parsing logic that allows users to perform queries on semi-structured data using the well-known SQL query language, Damany said. This is especially important, given the relatively low number of database professionals familiar with NoSQL queries.
Under the Radar
Founded in late 2011, BitYota principals raised $12 million in funding in 2012 and have spent much of the rest of their time purposely under the radar, Damany said.
"We have very deliberately not done a big bang marketing release," she said. "Data warehousing is a complicated technology, so we have spent the last three years creating the technology, understanding customer use cases and building partnerships."
The company has been building relationships with companies that will make it easier for BitYota customers to bring in data from disparate sources and visualize the results of their data analysis, and it plans to announce those partnerships soon, Damany said.
Fast Facts About BitYota
Founded: late 2011
Product: Data warehouse-as-a-service
Funding: $12 million, with investors including Globespan Capital Partners, Social+Capital Partners and Andreessen Horowitz.
HQ: Mountain View, Calif.
Ann All is the editor of Enterprise Apps Today and eSecurity Planet. She has covered business and technology for more than a decade, writing about everything from business intelligence to virtualization.