Splice Machine Moves Hadoop RDBMS into Public Beta

Splice Machine, provider of a Hadoop RDBMS, is moving its real-time management system from private beta to public beta, available for anyone to download and try.

The value proposition for the Splice Machine database, according to Monte Zweben, CEO and cofounder of Splice Machine, is that it enables companies to replace traditional RDBMSs when they hit a wall, either from a performance or cost perspective, with a full-featured, transactional SQL database on Hadoop, to power both operational applications and real-time analytics.

“Our go-to-market strategy is a freemium model where anyone can develop on Splice Machine and essentially do so for free, and when they put it into operation and derive value out of it, then they would license the platform on a per node basis, very similar to the Hadoop pricing schemes from the Hadoop distribution companies,” explained Zweben.

Companies come to Splice Machine either when their database is starting to get so much data from a volume perspective or the data is coming in so fast from a velocity perspective that they need to change and they need to evaluate their options - and typically in those situations, said Zweben, they are comparing two options – scaling up on specialized hardware from their existing database vendor or scaling out on commodity servers. The first feature Splice Machine provides is the ability to scale out, enabling them to grow from gigabytes to tens and hundreds of terabytes all the way through to petabytes by essentially scaling out commodity hardware, said Zweben, and, second, they don’t have to change their code.

As an example, Zweben described how Harte Hanks, a direct marketing services company, had leveraged Splice Machine. Harte Hanks has a software as a service for campaign management for retailers that they built on IBM Unica, a commercial off the shelf application. It also uses Cognos for reporting and Ab Initio for ETL processes.  When it needed an alternative to its existing RDBMS, the company looked for a solution that would not require it to change its application, said Zweben.  Harte Hanks was able to connect its Unica application, Cognos reporting and Ab Initio ETL directly up to Splice Machine because it is ODBC-compliant “and get a scale out version of their architecture to meet their performance-cost objective,” he said. “Since we are ANSI SQL, ODBC-compliant and transactional, they can simply replace their existing technology without retraining their folks and without changing their code.”

According to Zweben, “almost all the technology that you hear about both on Hadoop and on other distributed technologies are essentially analytic in nature only. We are uniquely bringing an RDBMS to the table that can handle transactional workloads as well.”

Splice Machine already has 15 charter customers across a variety of industries, including digital marketing, telecom, and high-tech. To download Splice Machine, visit