CenterBoard Pushes Business Intelligence To Deeper Levels

CenterBoard earlier this week launched CenterBoard Integrator, a unique approach to data mining that allows solution providers to rapidly build and deploy enterprise-class dashboards and business intelligence applications. CEO Thomas Henn and CenterBoard CTO, Dr. Julian Navez, developed the core technology as part of a battlefield asset tracking system for the first Persian Gulf War. Along with the product launch, CenterBoard has initiated a partner program, which features both charter and premier levels. In an interview with CRN Editor-in-Chief Michael Vizard, Henning talks about how CenterBoard is planning to change the fundamental dynamics of the business intelligence market, among the hottest categories in enterprise computing.

CRN: What does CenterBoard do exactly?

HENN: We position ourselves as being a virtual operational data store company. What that means is we have the ability to access information anywhere, anytime across the Internet. When people, as individuals do a search through the Google search engine, it comes back with a list of information and then you have to drill down even deeper. We actually take it a step further for business users so they can ask me specific questions and bring back the exact information that might reside in various databases.

CRN: How will this technology be used in concert with other applications?

HENN: Business intelligence applications, as an example, are an interesting market for us because customers typically have data that they either can't move or don't want to move -- but they do want to incorporate into their business intelligence applications. We allow them to do that very quickly. We give customers realtime access to information across and beyond the enterprise without replicating the data. And we do that with 90 percent less cost and 90 percent less time than the alternatives.

The alternatives for the most part are people moving data to a central location so they can query that information. That costs anywhere from $5 million to $10 million, takes literally months to a year to build and, if you want to add more data, takes weeks to months to accomplish. We can also effectively provide a single customer view across the enterprise. One of the biggest challenges with RFID is data integration down to the item level. We can help with that.

CRN: Given that, how does your system keep track of what data is where, and the relationship that the data has with other data in the enterprise?

HENN: We do automatic connections. We do dynamic data flow paths. There is no hard-coded work that you have to do for queries and we automate the profiling of the data. Those profiles provide a very compact way for us to represent whether the information is there or not. That's then tied into standard SQL representation so that when an application or BI tool sends out a question, it's really sending a SQL statement. We interpret that statement and route it to a specific location where the data is. If we have to unfold the join or the information because it's gone to multiple data sources, we will do that and then fold it back up and bring it back to that BI user as if it were going to a single database.

CRN: So the transport you're using, whether it's a database API or an XML Web service, is irrelevant?

HENN: You hit it on the nose. What we're offering is a much richer view of the information than any one of those choices because we give you access to all of those.

CRN: How did the core technology get developed?

HENN: Our CTO, Dr. Julio Navez, when he was doing his Ph.D. work at Rutgers, got DARPA funding for a project that dealt with battlefield asset management, for lack of a better description. They were having a problem because they gave each individual soldier a backpack with a GPS. But when they were sending the information across the network, it was literally doing a broadcast storm. Some of the information was more pertinent than other information. But more important was that the network just bogged down.

Julio came up with the idea of geographic-based routing, where you could circle information on a map on your screen and you would know that troops would be in that area where you wanted to send information. And that information would only go to those troops. You segment the data based upon geography. He then went to the Siemens incubation arm in Berkeley.

When he went to talk to enterprises they said geographic routing is good but really what we need is any type of content. So he decided to build a data management capability on top of network services. But instead of IP-based routing you now have content-based routing with the look and feel of a standard database. With the Internet, you have a network and it's indexed, meaning that you know the IP addresses. We've taken that a step further. You can think of us having built a database index on top of that.

CRN: Where does the channel play in this?

HENN: We are targeting for most of our business to go through the channel. We've signed up 18 to 20 integrators. Obviously most of them are focused on business intelligence and data integration.