Network Spaces: Revolution Computing’s Open Source Statistical Software

Subscribe via: RSS EMAIL

Categorized | LinuxToday, LinuxWorld, Open Source

Tags : Alfresco, BI, Compiere, CRM, Greenplum, Nat Friedman, Open Source, OpenBravo, Pentaho, REvolution Computing, Statistics, Venture Capital, Zenoss

Network Spaces: Revolution Computing’s Open Source Statistical Software

Posted on 23 January 2008

This week was a big week for open source software with funding announcements for a number of companies including Alfresco open source document management), which closed $9 million in series C. GreenPlum (open source BI data warehousing) pulled in $27 million. My own company Zenoss nailed $11 million in series B too. Somewhat lost in the shuffle was Intel’s investment in REvolution Computing creator of parallel computing software for computational statistics.

REvolution Computing provides software and support for high-performance statistical and business intelligence tools, including R. They provide commercialization and core parallel computing technologies (automatic parallelism in most cases) that renders R practical for production and regulated use. According to REvolution the ability to effectively deal with very large data sets and provide high-value research was reliant on expensive tools.

I am not a statistics wiz but I tried to dig into what they do, here’s what I got from their FAQ on SourceForge.net

Perform administrative tasks on a cluster
Run a large set of unit tests in parallel
Monitor activity on a cluster or set of workstations
Perform a statistical study using Monte Carlo techniques
Write a simple chat/talk program for use on your network
Prototype MPI programs
Write simple client/server program between Windows and Unix machines
Allow scripts to cooperatively lock resources, rather than using traditional lock files

The reason this caught my eye was that this is one of a number of companies that are moving into the mission critical applications space. I remember in the 2003 time frame that Novell used to give talks about how Linux and open source applications were being used on the edge of the network for file servers, DNS, web servers, mail servers, and the like.

I recall Novell’s Nat Friedman using a bulls-eye drawing that showed open source was penetrating the middleware and application layer but had a long way to get to the ERP, Business Intelligence (BI), and CRM applications that contained your most critical data ( the center of the bullseye). I think their analysis was accurate too, though I think we are already surprised by the speed which open source is becoming relevant in that most mission critical space.

I don’t think that they would have forecasted the velocity at which companies like Pentaho, OpenBravo, Compiere, GreenPlum and others would have made in roads to BI and ERP. I also think that SugarCRM is a great proof point in open source customer relationship management and to a lesser degree so is Concursive.

As more data is collected and warehoused in OSS apps it’s nice to see statisical tools to analyze that data though I suspect that REvolution is probably of just as valuable for industries that do a lot of number crunching like biotech and oil and gas.

No TweetBacks yet. (Be the first to Tweet this post)