Take Down These SaaS Walls

One of the more challenging issues facing IT organizations going into 2010 is trying to figure out how much to rely on software-as-a-service (SaaS) platforms to deliver applications versus running those applications on their own internal infrastructure.

But the real question IT organizations might want to ask themselves in 2010 is why they have to make this kind of decision in the first place. In an ideal world, IT organizations should be able to develop an application that can be dynamically deployed as a service or on their own local infrastructure as they see fit.

That’s what will ultimately differentiate various platform-as-a-service providers that in 2010 will all be vying to attract corporate developers to their cloud computing services. One of those platforms is from a company called LongJump, which competes with a host of cloud computing platforms that are all trying to expand the reach of SaaS applications. But as LongJump CEO Pankaj Malviya points out, very few of these platforms provide the flexibility that a corporate enterprise is going to need, especially in a new era of virtualization that enables application code to dynamically move anywhere across a network of virtual servers.

LongJump, says Malviya, provides that flexibility by allowing corporate developers to create either browser or rich client applications that can be initially deployed in a cloud, and then with a push of a button, re-hosted to run on local infrastructure. This approach gives IT organizations the flexibility they really need to be ultimately successful in a world where on-premise and SaaS applications are ultimately going to blend.

Malviya adds that IT organizations will increasingly need this type of capability to support the rise of “disposable” composite applications that end users will want to create, which in all probability will require moving data and application logic back and forth between local servers and clouds.

What's really interesting about all this is it’s pretty clear that by this time next year, the whole concept of SaaS could be moot as we move into a new phase of enterprise computing where the walls are defined by the needs of the end user, rather than the underlying architecture.