Cloud Principles - Service Models

Hello

This time I will try to simplify some concepts.

The other day I was presenting the cloud strategy for an organization to a CEO and I realized that most people at technical levels understand very well the basic cloud models, but for Business Decision Makers in this "as a Service" era this can be tricky. If we talk about IaaS (Infrastructure as a Service), PaaS (Platform as a Service), SaaS (Software as a Service) maybe they don't have yet a well understanding. I chose to write about those three core models, nowadays we have plenty of services offered in the cloud, the other day I even heard about new models like Bot as a Service.

There are some great illustrations that shows us who manages what in each model like the picture above, but something like that it's too much technical for a CEO. So that's why I decided to write this post and try to maintain the concepts simple.

Back to Basics

Let's start from the begining, what would you have right now to support the services offered by your organization?. Something like the following:

• Compute that consist of virtual machines (servers) supported by a virtualization infrastructure
• Network devices to allow connectivity between your infrastructure and
• Storage to keep all the critical information at a safe and centralized place

After that, we have the platform layer that includes all the components that will process the information coming from the infrastructure layer and transform it or add some capability or value to be tranfered to the upper layer

At the top we have the Software or Applications, this is the end-user facing layer where you offer a Service to the users, like a Web Site , E-mail or an internal application

Now that we have the basic concepts clear, let's translate them to the cloud.

Cloud Models

The three core models available on the cloud are:

• IaaS (Infrastructure as a Service): as the name explains for itself in this model you will have available the same components Compute, Network and Storage under a virtualized environment. So you won't have to deal with hardware anymore. You just need to validate how many compute/storage or network resources are required and provision them.
• PaaS (Platform as a Service): in this case we have this components at the higher level available, the most common in Microsoft Azure would be Databases and Web Apps (formerly Web Sites). In this case you just have to decide what kind of database or webiste you need, what's the desired capacity and create them with just a few clicks.
• SaaS (Software as a Service): This is the third core model of service offered, at this level you have the service with all the infrastructure and platform components already deployed for you. So you don't to worry abouth the installation and maintenance of those components, you just have to define how many users will use the service and start using it. Some examples would be an app deployed directly from the Azure Marketplace, the e-mail service of Exchange Online or a SharePoint site in Office 365.

Conclusion

Keep in mind that I only mentioned a few services, so you can easily relate them to each model. In Azure we have hundreds of services and features available and new features are added almost every week.

Microsoft is the only one Cloud Provider that leads the services offered accross the three models IaaS, PaaS and SaaS according to Gartner:

References

Microsoft announces industry's first Cloud Bot as a Service

Microsoft Azure Calculator

Microsoft: a Gartner cloud computing leader across IaaS,PaaS and SaaS

Microsoft Azure Services Directory

Microsoft Azure MarketPlace