Lavastorm Adopts R for Better Big Data Understanding

October 23, 2012 Alex Woodie

Among the slew of software vendors that for some reason chose last week to make big data announcements was Lavastorm Analytics, the developer of the versatile Lavastorm Analytics Platform. The company unveiled a new R Analytics Pack that enables the open source R statistical analysis programming language to work within the Lavastorm product. The company says the combination of R and Lavastorm’s visual approach of discovering and understanding data and processes will ease the challenges presented by “big data.”

The Lavastorm Analytics Platform offers an impressive array of features that spans reporting, ETL, SQL editing, cross referencing, data cleansing, and fraud detection. The software is basically used to help companies work through data and answer questions that can be tough to formulate, or “visual, discovery-based analytics” as the company puts it. It’s often used in “manage by exception” cases by large IT organizations with very complex business processes that touch a lot of different systems and have a large volume of data, such as Lavastorm’s parent company, Martin Dawes Systems, which runs an IBM i-based customer care and billing system and uses the Lavastorm tool in that environment.

You can add one more arrow to the Lavastorm quiver with the addition of an R analytical engine. The R Analytics Pack will allow Lavastorm customers to integrate analytic programs written in the open source R language into broader analytic business flows generated by Lavastorm. This will be useful for delivering hard-core statistical, financial, and predictive analyses of large data sets on top of the traditional business process analytics done within Lavastorm.

Mark Marinelli, vice president of technical product management for Lavastorm, sees the new R pack being used in a couple of different ways. For starters, R will provide statistical muscle to the “manage by exception” data management and business process integrity work that Lavastorm is so good at, he says.

“We haven’t included in our suite things like predictive analytics, so R really complements what we do very well there,” he says. “If you have discovered some interesting pattern in the data [using Lavastorm], and you wanted to do a regression and predict how that pattern is going to go forward, you bring R to bear as a complement to what we’ve done.”

For example, if a business is using Lavastorm to analyze the rate of uptake of a particular product by a customer over time, the analysts can use the new R pack to explore what the future may look like. “Our product will help you find out where you have infirmities in your business. What it won’t do is plot out the next three months,” Marinelli tells IT Jungle.

The second use case for R involves leveraging Lavastorm’s extract, transform, and load (ETL) capabilities. “If you’re working with data in R, or one of the packages associated with it, then you don’t have the facilities to do much more than pull in file or database connections. You can’t do the full-on ETL-plus analytics capability that we bring to bear,” Marinelli says. The R Analytics Pack gives users “a wider data footprint … and a more trustworthy foundation for whatever you’re going to do in R.”

R has blossomed as an open source product over the last several years. There’s a rich community of developers writing new statistical packages for R, many of which can be obtained freely over the Web.

The goal with Lavastorm’s new R pack is to take advantage of this rich R ecosystem, and to make it a repeatable part of the business process. Once the R program has been written by the statistical quants, other less-gifted (but still really smart) business analysts will be able to reuse them as “building blocks” within the Lavastorm framework. It’s point, click, then bask in the glow of super advanced statistical analysis.

“The trend now is to make investments in R-based analytics more accessible to business analysts, not just the programmers that created them, so that the analytics can be more broadly used across the business to analyze and control various operations, and Lavastorm’s unique visual interface will help this greatly,” says Carl Lehmann, a research manager at 451 Research.

The R Analytics Pack is available now as a free upgrade to Lavastorm customers already under maintenance. For more information, see the company’s website at www.lavastorm.com.