This series of posts is an Intel primer of our view of Cloud Services.
As vendors begin to stake out their claims in the cloud computing space it becomes important to clarify key terminology in order reduce ambiguity. We rely on two categories of definition, the services themselves and the architecture in which the services reside. First we will identify the three prevalent cloud-oriented architectures seen in the Enterprise today.
Cloud services are realized within three subcategories: software as a service (SaaS), platform as a Service (PaaS), and Infrastructure as a service (IaaS).
- SaaS is software deployed on a remotely hosted environment for supporting core business applications such as CRM, documents, “office” type applications and email. PaaS includes platform offerings, such as web hosting and application hosting.
- PaaS offerings are typically one level lower than a finished application, such as Microsoft Windows Azure, Google App Engine and Amazon Web Services. As a side note, most XML web services fall into this middle-ground category and some have invented an additional category, called Software Infrastructure as a Service (SIaaS) to capture services distinct from a specific SaaS vendor, but to keep things simple we will think of these simply as smaller platforms and bundle them within the PaaS category.
- Finally, IaaS refers to infrastructure itself, in most cases bare servers or virtual machine environments with just the base operating system installed. From the business perspective, cloud services support service-based billing and simplified payment (credit card or equivalent) with ubiquitous access over the Internet and specific, documented service-level agreements.
Cloud services also have specific abstracted technical attributes, which include massive scalability, multi-tenant architecture, and elastic use of the underlying physical hardware resources.
Next will be Cloud Architectures