Platform as a Service architecture

So far, we’ve have talked primarily about migrating applications in a fairly simplistic way; that is, by
simply copying VMs from an on-premises datacenter to the cloud provider—the IaaS model. Of
course, IaaS carries with it a number of advantages, such as passing responsibility for the datacenter
to the cloud provider. To really transform to a cloud-centric model, designing applications specifically for the cloud is the next step.

IaaS has certain limitations: you are still responsible for maintaining the system software, operating
system, and database for your application, including items such as periodic patches and software
upgrades. In fact, we can say that IaaS is only the first step in fully taking advantage of the cloud.
In Platform as a Service (PaaS) models, you only need to maintain your application, whereas the
system software is provided by and maintained by the cloud provider. In addition, PaaS offerings
typically add seamless scalability and resiliency by providing scale-out and data replication, and PaaS
can interact with cloud services such as Microsoft Azure Active Directory for robust identity
management.

Azure App Service Web Apps, for example, provides a way to rapidly provision a scalable website in
the cloud with a minimum of effort. Microsoft provides the underlying web infrastructure (operating
system, networking stack, storage, language support, and scalability features) that remove much of
the systems overhead of managing a large-scale web application. It is straightforward to configure scalability, backup, and monitoring capabilities into a Web Apps application. Web Apps also connects
to all the other services offered by the cloud for rich applications.

Azure Cloud Services are a cloud analog to the “three-tier” line of business applications of a decade
ago. In Cloud Services, an application consists of three components: a web role, effectively a web front end, scalable independently from other parts of the application; a worker role, providing background computation and processing (analogous to the business logic layer in the three-tier model); and persistent storage using an Azure-enabled version of SQL Server (Azure SQL Database). Although it requires some redesign to take an existing application to Cloud Services, this will be relatively straightforward because the model is intentionally similar to three-tier.

Source of Information : Microsoft Enterprise Cloud Strategy

No comments:

Cloud storage is for blocks too, not just files

One of the misconceptions about cloud storage is that it is only useful for storing files. This assumption comes from the popularity of file...