Feb

Cloud Computing and Windows Azure

Let’s talk a bit more about Cloud Computing….

Imagine an ASP.NET application installed on a group of servers for serving requests coming from web clients. Imagine that behind this battery of servers we have a group of machines for business rules, coordination of processes or services introduced in WCF – Windows Communication Foundation, providing functionality for this application.

Also found in the back-end machines- databases or systems integration with legacy enterprise applications accessing the so-called lines of business, or LOB – Line of Business Applications. This scenario is very common in business today, implemented in their own infrastructure (on-premise) or possibly with machinery suppliers and local hosts.

From a marketing campaign, which puts company in evidence, or even because of the growing interest in application functionality, the transaction volume is expected to grow over time, increasing the number of requests or even concurrent users and pages Solution Services. In your planning, you can prepare for this growth, triggering buying more dedicated hosting servers, more network infrastructure and configuration of the new part of the IT application. With this growth, we gained an additional cost of operation, more machines to manage, monitor, while you become responsible for updating system software and installation of the application functionality on more machines. The risks and challenges increase, it is not true?

Now imagine that the volume of access requests decrease for some reason, stabilizing at a level less than the amount originally planned for growth, going back to the original volume of access. The machines are now idle most of the time, while consuming operation costs, administration, power, monitoring, etc. It would be interesting to simply change a configuration file and “deprovisioning” these additional machines, paying back the original cost of the system? How about doing this provisioning as many times as you want, as long as necessary?

The scenario presented is a perfect model for cloud computing and Windows Azure.

Cloud Hosting:

Simply, we can define cloud computing or cloud hosting as an environment for processing and storing massive data, high scalability and high availability, accessible via web interfaces such as HTTP, REST and SOAP, installed in data center around the latest generation world. A great feature of this model is the provisioning of elastic computing: looking at the example above, we can start the operation of the system with 10 machines planned for the front-end and back end.

With the increasing volume of requests, we can hire new machines, moving to a scenario of 20 machines to the front-end and back-end, dynamically, by setting the environment via an administration portal, for example. After the peak usage, we can simply return to the original configuration of 10 machines, just updating the configuration of the environment in the cloud.

This power configuration is known as elastic computing, where provisioning is done dynamically according to application demand. At the same time, new models of contraction, subscription services and associated licensing, where the company can only pay for the infrastructure that it actually uses.

Doing a little math, you’ll see that the scenario of cloud computing may offer a significant cost reduction in administration and operation of IT in business, which has accounted for the increased interest of customers and software vendors about the model, as Microsoft itself .

An operating system in the cloud

Windows Azure is an operating system that runs in various data centers worldwide. Among the features found in Windows Azure highlight its ability for “utility computing”, where various services are offered by the operating system to manage and abstract features present in the data center, including:

Easy to expand to new locations (geo-distribution);
Live update of new software features;
Applying patches and fixes operating system;
Diagnosis and recovery from hardware failures in their own data center;
Large storage capacity with high availability;
Dealing with traffic growth;
Diagnose and respond to service failures;

Some application scenarios are particularly adherent to the cloud computing model, such as:

SaaS LOB Applications, or lines of business applications on SaaS – Software as a Services offering features. Some examples are SaaS CRM, SaaS HR, ERP SaaS, etc.
Collaborative Web Applications involving multiple devices and users with the handling of different data formats.
Hub applications with shared data, involving a growing volume of data shared among several companies, such as supply chain scenarios, integration of businesses, etc.
Web Applications with Business Intelligence and Data Mining, with increasing volumes of data and information storage;

Cloud services, cloud processes, data in the cloud, on-premise/off-premise collaboration, interoperability with other platforms in the cloud and a series of tools is used for developing applications.