RainStor unveils native Hadoop database

Jennifer Scott News
18 Jan, 2012

With its patented compression technology, the UK firm claims to have the best solution for analysing Big Data.

RainStor today launched Big Data Analytics on Hadoop, claiming to offer the first enterprise class database native to the Apache project.

Aimed at large enterprise deployments – especially telcos and financial institutions – RainStor used Hadoop as a platform to build a database on; including its new APIs but incorporating tools from SQL, ODBC (Open Database Connectivity) and JDBC (Java Database Connectivity).

The idea is to allow the scalability and Big Data analytics provided by Hadoop, whilst using the traditional skills of database engineers.

“If you have got 25 years experience using Oracle, you will have the ability to apply your engineer,” John Bantleman, chief executive (CEO) of RainStor, told Virtual Clouds.

“You will be able operate at scale, but leverage the technology and skills that you already have. That is different to the products out there; it doesn’t exist.”

Another factor Bantleman said differentiated RainStor’s offering was its patented compression technology. He claimed the algorithm could reduce the size of the information by up to 60 times, turning 1PB into just 20TB of storage.

On top of the reduction, Bantleman also said users would not have to “blow the data back up” to analyse the raw data. Instead the analytics can run on the compressed information, with the addition of a management layer.

“The basis of our compression is store everything and repeat nothing,” he said. “The disk is smaller, IO is smaller, memory is smaller; your whole cluster operates faster and simpler.”

“We have compared ourselves to other offerings and most people are popping champagne corks at 2X or 3X compression, we move that by a decimal point with how well we store data.”

Despite being a six-year project by the Apache Foundation, Hadoop only reached version 1.0 at the beginning of the year and Bantleman admitted his firm were dealing with newcomers.  

“I think customers are still learning, and everyone is still learning the technology,” he said. “There has been so much [noise] around Hadoop as people are starting to say ‘I have the peta-scale problem’ and they are thinking of the economics, as well as alternative architectures.”

“Over the last 12 months people have gone from ‘I want to run this on a scale-out NAS’ or ‘I want to run this on a cloud type to environment’ to ‘I want to use a Yahoo-type cloud or Apache Hadoop architecture.’ So I think we are probably doing first deployments with those companies.”

Big Data Analytics on Hadoop is available globally today, either directly from RainStor or through one of its partners – including Dell.  

Sign up for our free newsletter