In-Depth

BI Trends: IBM’s DB2 9 Release Answers Questions, Begs Answers

BI junkies must wait until later this year for IBM to shed more light on its DB2 9 BI strategy.

• By Stephen Swoyer
• 06/14/2006

The DB2 version 9 release IBM Corp. announced last week helped answer a few long-simmering questions about Big Blue’s competitive strategy vis-à-vis Microsoft Corp. and Oracle Corp. on the relational database front. However, as far as business intelligence (BI) features are concerned, DB2 9 isn’t all that different from its predecessor. BI junkies will have to wait until later this year for IBM to shed a bit more light on its RDBMS-specific BI strategy.

The newest version of Big Blue’s flagship database does boast native XML support, which IBM officials say helps make the revamped DB2 the industry’s most XSLT-, XPath-, and XQuery-friendly RDBMS on the market.

“As an application developer, you can write a query intermixing SQL, XPath, or the new XQuery language, and DB2 does the work for you to query both, any and all data sets in the database, whether it’s the relational structure or the [XML] data structure. So you the developer don’t have to write the extra code or take the extra time,” comments Bernie Spang, director of data server marketing with IBM’s information management group.

Spang says IBM’s XML implementation is superior to those of Microsoft and Oracle because Big Blue stores XML as XML—“in its pure data structure”—inside the database. Microsoft and Oracle, Spang maintains, use techniques like XML “shredding” (breaking up XML information and turning it into rows in relational tables, as Microsoft did in SQL Server 2000), or store XML information as Binary Large Objects (BLOB) or Character Large Objects (CLOB) in the RDBMS itself—which means they effectively dump XML information en masse into relational tables. “The problem with that is that if you want to gain any insight, you have to pull the whole CLOB out and then you and your application code have to decompose it, or you have to have some intermediate layer to parse it and do the query,” he argues.

Not surprisingly, IBM’s competitors take issue with this claim. Microsoft, for example, championed native XML support as one of the strongest selling points of its SQL Server 2005 database, and the thoroughgoing native-ness of SQL Server 2005’s XML implementation became a cause for some concern among SQL programmers, who feared that other, non-SQL programmers would inevitably abuse it—thus compromising the integrity or reliability of SQL Server data. (http://www.adtmag.com/article.aspx?id=11148&page)

On his blog, Michael Rys, program manager for Microsoft’s SQL Server Engine Team, has attacked IBM’s claims.

“Both DB2 and SQL Server (and others) expose or will expose at the logical level an XML datatype that provides XML fidelity plus query and update functionality. Thus all of them provide ‘native’ XML capabilities (without abusing the language),” Rys wrote on his Weblog last year. “IBM's physical design is irrelevant. Whether you store it as a string, store it in some internal binary format making use of existing storage facilities provided by the relational database system or design a complete new storage engine does not matter.”

DB2’s BI Feature Set Missing in Action

Notwithstanding its much improved XML support story, the vanilla version of DB2 9 Big Blue announced last week doesn’t include much else in the way of new or enhanced BI goodies—at least relative to what competitive offerings from Microsoft, and, to some extent, Oracle have on tap.

Oracle’s 10g R2 RDBMS includes OLAP and data mining capabilities, for starters, and Oracle continues to prep an ambitious redesign of its next-gen Warehouse Builder ETL tool (http://www.tdwi.org/News/display.aspx?id=7636). Microsoft, for its part, recently shipped its SQL Server 2005 database, which substantially refreshes its SQL Server-based BI stack. The revamped SQL Server includes an enhanced reporting component (Reporting Services), a thoroughly redesigned ETL facility (Integration Services, which replaces the erstwhile Data Transformation Services), and boasts an improved OLAP and data mining engine (Analysis Services), too.

IBM does provide basic ETL capabilities in DB2, via programmatic SQL. But Big Blue stepped up into the data integration big leagues early last year, when it acquired the former Ascential Software Corp. Some industry watchers have suggested it would behoove IBM to expose a limited subset of Ascential’s ETL capabilities in DB2, thus giving that database feature-parity of a sort with its competitors. Last week’s DB2 9 release was a no-go in that department, however. If anything, IBM officials said, BI junkies must wait for the version 9 release of DB2 Warehouse Edition, which is slated to ship later this year.

“The first level of additional data warehouse tools are the additional capabilities … [we plan to deliver] as part of DB2 Data Warehouse Edition 9,” Spang confirms. “We’ll be announcing the Viper [DB2 9’s codename] update to DB2 Data Warehouse Edition later this year, so stay tuned.”