IBM WebSphere

Released last year, WebSphere App Server V5 was a significant upgrade for IBM. In a succession of previous releases, the company had quickly pulled together a variety of products. With V5, IBM could start to tout WebSphere as working off of a single code base, albeit with multiple configuration options. Formal availability was announced in November 2002.

The release was also marked by support for J2EE 1.3 Message Beans and IBM's native high-performance Java Messaging Service (JMS) implementation. V5 also supports container-managed messaging. Like others among the top rung of J2EE app server makers, IBM is increasing its focus on performance scalability and app server management, especially as it relates to reliability and availability.

INTRO
IBM
BEA
SUN
ORACLE
MACROMEDIA
MORE PLAYERS
SINGLE PAGE VIEW

Version 5 includes improvements that allow administrators to use Java Management Extensions (JMX) that record and log stats on usage and resources [see Manage Java apps for premium performance , by Peter Bochner]. Support like that is a clear sign that Java is moving toward wider, everyday enterprise deployment.

IBM posits its administrator-related improvements under the aegis of its autonomic computing initiative. The system can be set to automatically detect and resolve, some problems. This is described as a form of automatic tuning, and is associated with system self-healing and self-optimization.

''The application server has become an operating system for the network,'' said Scott Hebner, IBM's director of marketing for WebSphere. ''Now transactional capabilities become real important,'' he said.

''With WebSphere Application Server 5, we are seeing a shift in usage,'' noted Hebner. ''It's much more than just a J2EE server. It has a full complement of business integration facilities and integrated workflow. It supports clustering and multi-domain managing of server farms.

''The market is evolving into IBM's sweet spot -- high transaction environments,'' he suggested.

Looking ahead, Hebner foresees improvements in what people now call the ''choreography'' of various integrated apps. He also suggests that more advanced clustering and failover server traits will lead the way toward Grid computing, a notion forwarded by IBM and others.

Back | Next

INTRO | IBM | BEA | SUN | ORACLE | MACROMEDIA | MORE PLAYERS