Raj Nathan, senior vice president, Information Technology Solutions Group at Sybase, has always been a technology pioneer, but one who also likes to stay close to the customer. As Sybase has evolved, Nathan is helping guide the company to serve new customer requirements, which led to this recent conversation with DM Review editorial director Jim Ericson.

DMR: How has Sybase evolved to address changes in the marketplace over the last few years?

RN: In the old days, our focus was very database-centric. As we evolved, we began to handle the movement of data from [multiple] repositories and databases and looked at the uses of the data, the mechanisms of delivery and what [different] types of data need to be delivered. We are still very much focused on the management and delivery of data for operational tasks or decision-making. But today we've taken data used for transactional applications and automation of manual operations and applied it to other purposes - decision-making or remote access for the mobile worker. This also includes movement of data and messages to nonhuman forms like an application or a device. From purely relational table data we've come to address data that's stored in file systems, objects, nontextual data and textual message-type data. With the acquisition of Mobile 365, we are addressing the movement of messages and handling of data code between repositories, sources and consumers of data. So we are still centered around data in its many forms, but where our old boundaries were drawn by our repository, our central focus today is the data itself and the way the data is used.

DMR: As an infrastructure provider, how do you communicate the business value of Sybase to end users who never see the product?

RN: At a high level, the people with responsibility for business are asked to manage and improve margins for the company. Because it is a competitive environment and margins are getting tighter, we need to be more aware of the cost of operations and identify the things that support revenue. There are two elements: the things you need to do to support revenue, and the things you need to do to manage costs and not waste resources. Both of these functions require a lot of data and metrics so you can figure out what a marketing campaign is actually costing you, the result of the campaign and the cost per lead. With these kinds of metrics you can figure out whether putting ads on Google is more valuable than running seminars in several cities or placing an ad in a trade journal. I might want to know precisely which segment of customers are buying my product with an optional feature and why are they buying that option. This lets me identify market segments and go after other segments with the same characteristics. We need a more granular level of information to compete in today's slow-growth environment. This notion applies to just about every industry and because of that, more defined data, more granular data, more consumable data is needed for managers to make decisions.

DMR: Are the IT organizations of your prospects and customers selling this proposition internally?

RN: Sybase sells infrastructure technology, and so many times IT people will look at the technology and see how it could solve a business problem. As you know, the underlying technology doesn't much matter to a business user. The ability of the technology to deliver the results they want at the cost they want in the short term and over the long run is what matters to them. Once we establish that, we do get their attention and the business and IT organizations are helping each other. We support this on our Web site and in customer interactions with use cases. The fact is, these technologies are of such a magnitude that just purely understanding the technology without the use case will not produce results.

DMR: You launched your data integration suite in August. What does that represent in terms of an improvement for your customers?

RN: We've been building the data integration part of our business for eight or nine years now in piece parts for our replication server, which focuses on data movement services, data distribution services or data integration. We have products for mainframe data access, and other technologies that can look at other databases as if they are part of our own database. We've developed and acquired quite a few technology pieces, not necessarily integrated, and if you look at trying to get data from different repositories, staging the data, presenting multiple views to multiple customers, we can clearly see that no one data integration technique will cover it end to end. It is very conceivable that you will use different technologies for different phases of a project - in the staging area you might use ETL technologies, and a different technology to distribute it to end users and provide ad hoc queries. Rather than the customer having to worry about how all this works, we decided to provide a platform where all these services become available. The individual piece parts and services have been one of our highest growth areas for us over the last 12 months. I don't see it any different with the suite, but the difference is, the management console and underlying infrastructure is all the same. You can buy a server product and choose only the service you need. If you want to use the same server for another service, you can just license exactly what you need. Going forward, Sybase will provide granular services in an integrated whole, a base framework that is integrated but doesn't leave you with shelfware. You don't want or need to reinvest in new frameworks, monitoring or management tools.

DMR: How does data integration fit into your overall strategy?

RN: Our core business continues to be the database where we have a very good and cost-effective product. Separately, we have released segment after segment for the mobility suite, iAnywhere, the mobile data stores; all that broad mobility business is becoming close to a $200 million business for us. We are the number one market share leader - close to 70 percent - with the ASA mobility database.

DMR: How and why has the mobility strategy grown?

RN: Here again, we have extended the database-centric heritage to the middleware component and other infrastructure technologies that are related to data. Beyond the data store, we now do device management and security on devices. It is important because we see an opportunity to expand our leadership position as the whole mobility market is expanding. Our traditional mobility business was focused on the enterprise customer, the field sales and service force and data connection for branch automation or RFID devices. With the acquisition of Mobile 365 we also get into opportunities with telco carriers, and through that we get into the consumer space. Today, the portable business user doesn't carry two portables; they blend Google searches with the things they traditionally do on the job. Mobile 365 helps us because we think, in the long run, the customer and enterprise mobility markets become the same. Another point that is slowly coming up - we don't have hundreds of examples of this - we do have a customer that is accessing BusinessObjects by mobile device. This kind of usage could be for middle or senior management; they can look up key performance indicators, the 10 major deals in the pipeline that need to be signed or the last 10 deals that were signed.

DMR: Where does Sybase fit into the master data management picture?

RN: In master data management the issue is often not the technology, but the internal processes that have to be changed to be able to handle them. From a data integration perspective, metadata management becomes quite important. Customers have realized that the context should be the data - and not just not at the vertical dimension. There's also the problem of the horizontal dimensions. Customers have realized that you can solve the complexities of data integration to some degree by tagging it at the master data level itself. This is our context: A packaged application vendor tells a customer they need to upgrade to the next version. The issue is that 10 other business units have written 80 or 100 applications based on an old version that are going to be very difficult to change. There is a simple case for a layer of metadata structured between the applications, data schemas and what is visible to the downstream applications. Now the ability to upgrade the back-end application is very easy; you just make the connections between the new data schema on the master data. This is not an enterprise-wide metadata architecture or master data management; while that gets to be a much more grand thing to do, the right thing to do, it's a long process. There are a lot of immediate practical problems where an abstraction layer of metadata management is very useful.

DMR: So any solution needs to be viewed in the context of the problem at hand?

RN: People need relevant information to make decisions, and sometimes that requires very precise and comprehensive information, no doubt. And sometimes it is more important for us to get approximate answers in real time than it is to get precise answers delayed. Traditionally, you asked a database a question, it would look at all trillion rows and give you an answer. In the past, we would initiate a business event and either create the data or initiate the business event and ask for data. We are now getting into a stage where the changes in data predicate a business event to take place; I need to know up front what is happening and start an event. If my DSO [days sales outstanding] is going out of range, I need not only the alert, but also the top 15 accounts that are the most delayed or have the largest debts for the longest period of time. Those kinds of data change driven events will become more necessary. It will not be me coming in every morning and asking about DSO today; it will show the top 15 and warn that if they don't pay in the next week, the DSO will be in the red range. Before you get to the red range, you might as well go talk to those 15 customers and get the payment in place. That kind of intelligence driven by cognition of what data changes are taking place, these things are a little different from the past when data was principally about transactional applications and automating a manual operation. 

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access