Heath Brownsworth

Since Salesforce.com’s acquisition of Heroku in December 2010, the interest in Platform as a Service, commonly referred to as PaaS, has been steadily growing. January brought Elastic Beanstalk from Amazon Web Services. In April, VMware introduced Cloud Foundry, and just last week Red Hat announced OpenShift. These recent entrants join Microsoft Azure, which moved into the PaaS space in late 2009.

What’s driving all the momentum? On the surface it seems that no one wants to be left out of the game, but looking deeper it appears we’re simply witnessing the logical evolution of the cloud. Many of the first applications in the cloud had lives before the cloud. As such, their architecture wasn’t necessarily designed to leverage the full range of capabilities enabled by cloud computing.

Platform as a Service offers developers the opportunity to focus on the feature set of their applications without having to worry about the underlying infrastructure. Critical cloud computing elements such as provisioning, scaling and monitoring are built into the platform. Using PaaS to develop a new application is a foregone conclusion, but what about all those pre-cloud applications? This brings us back to all the new PaaS players.

To move forward, many applications will just need to be rewritten. And you guessed it … everyone wants to be your platform/cloud. While PaaS brings familiar names, the issues are familiar, too —  proprietary vs. open, efficient vs. portable, isolated vs. adopted. Even though PaaS adoption to date has been anemic, there’s little question that it’s the next big step for cloud.

Heath-