San Francisco - Gartner Symposium/ITxpo -- In the old days, if you were an IT shop or an application developer, you had almost total control over the environment in which your application lived. You had some code and it may have accessed a database or some other sort of structured data but it all lived on one system that you had control of. Even if that application had to venture out onto a local area network, that network was usually yours too. Back then, the number of things that could screw up your application was relatively small, and, although keeping things running was a complex task, it was usually something that one or two people could get their heads around.
Today however is a brave new world. Application code lives in one place, while the data lives in another. In fact, the data that an application acts on -- for example a Web application --is often distributed, living in several places, not just one. Not only don’t IT shops have control over the environments that their applications often interact in, they don’t control over most of the infrastructure that their transactions touch. For example, the Internet. To make matters worse, it isn’t only employees who are accessing these applications from 3270 terminals anymore. Now, business partners, stockholders, and customers need access, too, and from all different types of devices and applications on their ends. Simply put, keeping tabs on this composite class of applications is no longer humanly possible. Even if humans could scale to stay on top of the complexity of a given application’s environment, they couldn't possibly keep tabs on that environment in real time.
While at Gartner Symposium/ITxpo 2005, I came across Wily Technologies, a company that fills that niche. Priced at roughly $6250 per CPU (based on the number of CPUs your application is running on, not the number of CPUs that Wily's solution is running on), Intrascope is designed to keep tabs on every dance step your transactional J2EE applications take. The ROI of an averted government-mposed fine or SLA failure-induced refund alone could make the product pay for itself with one incident. If something goes wrong, not only can it alert you, it can also interact with autonomic management technologies like IBM's Tivoli that can automatically respond by taking whatever corrective action is necessary.
In my interview with Wily’s Senior Vice President of Marketing (available as an MP3 that can be downloaded or, if you’re already subscribed to ZDNet’s IT Matters series of audio podcasts, it will show up on your system or MP3 player automatically. See ZDNet’s podcasts: How to tune in), Mike Malloy talks about the sorts of Fortune 1000 customers that have come to rely on his company's technology, and why --because it's written in Java -- the technology can service any J2EE environment. (I ask whether .Net support is on the way.) Malloy also discusses how Wily plans to price its products as the industry trends towards multicore processors, and what the folks at his company thought about IBM's acquistion of open source J2EE provider Gluecode.