Virtualization & Cloud Computing

Hi folks.

On today's post we're going to analyze two increasing tendencies around data center design: Virtualization & Cloud Computing.

Nowadays the evolution process on all areas is amazingly fast, once something is adopted for production, it rapidly becomes deprecated. That is even more evident in the technological world. Hence, a data center design which takes only in consideration dedicated physical equipment is no longer viable.

Through the use of virtualization we can overcome the need  for specific services by virtualizing those in general purpose servers which provide the resources required. Virtualization per se is the creation of a virtual -- rather than actual -- version of something, such as an operating system , a server, a storage device or network resources. Therefore, we can create any kind of virtual device to fulfill any need the CPD may have without the need to acquire dedicated hardware, which would otherwise be hard to amortize.

Another strategy growing in use to reduce the dependency on actual hardware, specially on peak hours, is cloud computing. Cloud computing enables companies to consume a compute resource, such as a virtual machine, storage or an application as a utility rather than having to build and maintain computing infrastructures in house. Through the use of cloud computing a data center can be designed to hold a stable amount of workload instead of dimensioned to hold peak workloads. Then, a data center could rely on cloud computing solutions to handle all the workload their actual physical datacenter cannot. Another interesting use case is for high availability, redundancy and backup strategies. Storing and deviating some computational power to the cloud decentralizes the data center hence making it more robust and flexible.

If we put these two technologies side by side we can see a clear difference between them. While virtualization only attempts to  detach functionality from hardware, hence allowing for infinite functionalities only restricted by the resources they consume; cloud computing goes a step beyond that, it also allows for that containerized services or functions to be delegated to external resource pools so you can run them without the need to actually own the required resources.