Cloud Computing is the new hype, a term which is actually quite ambiguous.
Wikipedia definition is : Cloud computing is Internet-based computing, whereby shared resources, software and information are provided to computers and other devices on-demand, like the electricity grid.
It’s a very generic term, which comprises different services (infrastructure, applications) and typologies (public clouds, private clouds) but it’s definitely a booming sector.
In this post I just wanted to clarify the cloud service layering.
In the first picture, on the right, the usual enterprise configuration with its own IT: data, applications, development and infrastructure are in-house.
In picture 2, the configuration where the infrastructure is in the cloud, IaaS = Infrastructure as a Service. This can include the servers, the network and the databases but also system and security software.
Amazon Web Services is an example of company offering this service.
This is useful if a company is short of space or doesn’t want to own and maintain hardware itself (and the operational costs associated). Also useful if the demand fluctuates. For example, if you’re a tickets online selling company with an average demand but sometimes you are highly requested concerts as for the U2.
In picture 3 the configuration where the development and deployment environment is in the cloud, PaaS = Platform as a Service. This can include all the development tools and engines, information and data management software, integration and process automation, enterprise portals.
Windows Azure and Google Apps are example of PaaS.
This is useful for on-demand activities like prototype, test, pilot. For example for a pharmaceutical company. Also useful for fluctuating demands.
Finally, in picture 4 the configuration where also the applications are in the cloud, the SaaS = Software as a Service.
This includes any kind of applications, for example ERP and CRM software.
Salesforce.com and Oracle on demand are examples of SaaS.
This is useful for many applications, typically for ERP and CRM.