Software-as-a-service is at the forefront of the cloud computing trend. Successful implementations of applications in the SaaS model have blazed a trail to the cloud. But is every application potentially a cloud application? To flip it around, are there certain applications that inherently do not lend themselves to the cloud computing paradigm? If so, maybe we ought to figure those out now before someone gets burned.

It's possible that the distributed, decentralized, network-oriented nature of cloud computing may not lend itself to all areas of application processing. In a recent column, Chris Murphy, editor of InformationWeek, says there are many applications that don't fit the cloud model, such as "graphics-intensive apps like computer-aided design and many financial and transactional apps…"

Fair enough. Murphy points out that in an InformationWeek Analytics 2010 survey respondents cited "no business requirement" as the top reason for not implementing SaaS (actually, it tied with "concerns over security" as the number one reason). However, in a recent column, Alex Wolfe, editor-in-chief of InformationWeek.com, describes how Microsoft is working to make high-performance computing capabilities available in the cloud through its Azure cloud computing service. These types of capabilities may make the cloud a realistic platform for high-performance apps like financial and transactional apps. Of course, security is the elephant in the room whenever discussing cloud computing. That dictates a measured approach to IT strategy, especially in connection with sensitive, competitive, customer-oriented applications and something as new as cloud computing. But just because certain apps aren't appropriate now for the cloud, doesn't mean they won't be in the (near) future. And just because certain IT organizations don't see a requirement now for SaaS capabilities doesn't mean their competitors won't figure out a way to exploit those capabilities, and sooner rather than later.Software-as-a-service is at the forefront of the cloud computing trend. Successful implementations of applications in the SaaS model have blazed a trail to the cloud. But is every application potentially a cloud application? To flip it around, are there certain applications that inherently do not lend themselves to the cloud computing paradigm? If so, maybe we ought to figure those out now before someone gets burned.