This week’s major acquisition announcements of Bea by Oracle and MySQL by Sun are considered significant milestone in consolidating J2EE application servers and open source database servers. The application server market is witnessing a strong convergence with 9 Java EE 5 compatible implementations vs 17 J2EE 1.4 vs 22 J2EE 1.3. The players are WebLogic Server (BEA), WebSphere Application Server (IBM), OC4J (Oracle), Sun Java System Application Server (Sun), NetWeaver (SAP>), GlassFish (Sun), Geronimo (ASF/IBM), JEUS (TmaxSoft), Apusic (Kingdee). There is a decent choice for customers who either opt for commercial products or use the open source versions. This number can be further reduced to just 5 popular offerings, as Sun extends GlassFish, IBM extends Geronimo and Bea and Oracle is going to be one or two depending on Oracle’s new strategy. For the sake of this discussion, I am not considering other popular open source and commercial J2EE implementations which has a huge deployment base.
Historically, such acquisitions were not so successful at least for app server vendors : HP acquiring Bluestone, Novell acquiring SilverStream, Adobe acquiring Macromedia (Allaire JRun), Borland (BORL) acquiring TogetherSoft, RedHat acquiring JBoss. It will be reasonable to compare RedHat vs JBoss with Sun vs MySQL acquisition, because they both acquired most popular open source products. While it is interesting to see JBoss lacking momentum after its RedHat acquisition, this may not be the case with MySQL’s future and still be a successful open source product in the hands of Sun. All we know for fact that Sun does not earn much of its penny from Java, JEE products and its recent efforts in open sourcing various products including the Java language has proven its commitment to the open source community. This is a strong message to the open source community. So, don’t be surprised by similar acquistions in the future. Did I say framework vendors?
Oracle’s acquisition may rewrite this history, as OC4J is not so popular than its competitors. So, Oracle may just provide drop-in-replacement using WebLogic in its Fusion middleware stack. But, there is whole lot of Bea products which complements Oracle’s offerings. For instance, Oracle will benefit by having its own JVM (JRockit), TP Monitor (Tuxedo), Integration Server (WLI), and its huge installation base. So, the compelling truth is that Oracle is aligning its forces to compete with IBM. Sun was never a tough competitor in any of these markets. Just to balance the equation with IBM, Oracle is missing its own OS platform.
App servers are becoming commodity software, so why would one pay for it?