IBM hopes to strike hard and fast with its next database software, which features more native XML tools to help customers better tap into data housed throughout a service-oriented architecture (SOA).
Just released to customers, developers and partners for open beta testing, DB2 Viper is the first database crafted to integrate and manage native XML data and relational data.
This means XML data won’t have to be reformatted or placed into a large object within the database, allowing database administrators to increase the speed at which they extract data stored on a network made of disparate architectures.
This also means developers won’t have to create separate applications to access both relational data and XML repositories, said Bernie Spang, director of databases for IBM. Spang said one of Viper’s goals is to liberate data from the static form it has been forced into by relational database products.
Spang said IBM anticipates Viper will resonate well with companies who have or are thinking about implementing distributed computing models such as SOAs “This is critical as our clients and partners evolve into a service-oriented architecture approach using Web services,” Spang said in an interview. “Obviously, with the explosion of XML information in e-forms via Web apps due to document standards, it is becoming increasingly critical in the database to provide information as a service.” How does this new technological approach work? Instead of storing XML as a file or a blob in a cell in a database, Spang said admins can store and navigate among XML in its structure, allowing them to run analyses against the XML data similar to the way they could analyze columns and rows of relational data. Oracle and Microsoft currently lack this feature, although the top three database vendors all support some degree of native XML. IBM introduced its first native XML capabilities in its last DB2 launch, code-named Stinger. That DB2 version focused on publishing information in XML and providing it through a Web service interface. By adding the ability to store XML-based messages, such as stock transactions, in Viper, IBM hopes to put a hurting on Microsoft and Oracle in the database market, which Gartner Dataquest said is at $7.8 billion annually. Spang said adding advanced native XML capabilities can only help this, as customers seek easier ways to pull information from a database, which is essentially a big vat of digital information. DB2 Viper will also be the first database to support the three common methods of database partitioning at the same time, simultaneously handling range partitioning, multi-dimensional clustering and hashing. These functions will let admins arrange and order their information the best way for their businesses. Viper will also support XQuery, which is quickly becoming an industry standard language for processing XML data. Applications can use XQuery, standard SQL or both to retrieve documents from both storage formats. Spang also confirmed that IBM plans to extend early support of Viper to the PHP development community using Zend Core for IBM. One of the most popular Web programming languages in the world, PHP is used in over 23 million domains. IBM is offering a sneak peak of DB2 Viper this week at the XML 2005 Conference in Atlanta. The finished product will appear next year. Burton Group analyst Peter O’Kelly said XML support in database management systems is going to be a key trend for 2006, with IBM, Microsoft and Oracle all adding support for facets of XML storage, along with support for XQuery. He also said the new capabilities could very well put IBM in a nice position versus the competition heading into the new year. “Overall, I think we’re heading into a new phase in DBMS competition, with XML storage, XQuery, support for consuming and producing Web services, message broker capabilities, and more,” O’Kelly said. “The DBMS competition among IBM, Microsoft, and Oracle is going to be increasingly intense, and that’s going to produce some new and compelling opportunities forDBMS customers.”
