ARCHITECTING THE CLOUD: DESIGN DECISIONS FOR CLOUD COMPUTING SERVICE MODELS (SaaS, PaaS, AND IaaS)

Abstract

In the summer of 2008, after three decades of building software in corporate data centers, I walked away from corporate America to take a stab at building a technology company from the ground up based on a clever concept of the start-up company ’s founder. After years of building software within the constraints of existing data centers and the long procurement cycles required to make new computing resources available to build on, I saw leveraging cloud computing as an opportunity to achieve far greater agility at a pay-as-you-go utility pricing model. When I started my journey I tweeted to my social network and asked if anyone knew of any real-life examples of real-time transaction processing occurring in the public cloud. My tweet generated a lot of laughs and snarky comments; after all, who would have thought of processing information from a brick-and-mortar retail point-of-sale system over the Internet with a transaction engine in the public cloud in 2008? One responder laughed and said, “Let me know when you fi nd an example.” It was clear to me that we were pioneers, and we would have to learn things the way pioneers learned: by trial and error. Now, fi ve years later, I want to share my lessons learned with readers so they can rely more on the experiences of others rather than brute force trial and error, like pioneers.