K Data Grid

Arthur Cole recently spoke with Jake Sorofman, vice president of marketing, rPath, regarding virtualization and cloud computing.

Cole: Ubuntu specializes in application virtualization, among other things. How does it extend the benefits of virtual platforms already up?

Sorofman: Virtualization is part of the reinvention, to be held in the company because the economy is broken. The model business no longer works. The concept of application virtualization is essentially a way to deploy applications in a virtual fashion. Applications run on hypervisors and moved in a highly transparent environment. Folks Development applications are used to create functional software and throwing it over the wall for people operations. It is a laborious process of manually link software with the hardware, optimize and finally certification. This process may take four to six months. But today's business environment requires a deployment Quick and virtualization takes the friction process.

Cole: But you do not describe more of a middle cloud, rather a purely virtual?

Sorofman: Virtualisation is the foundation of the idea of cloud computing. The cloud adds a dimension of elasticity. While the previous model required investment in capacity and server management application workloads to meet peak demand, the elastic model calculation requires that you pay only what you eat. Applications can be provisioned, scaled, scalers, available upon request, and paid on a utility.

Cloud computing is the convergence of a number of trends: virtualization, SaaS, utility and grid computing. Ultimately, as these mature and are all gathered, the cloud becomes an interconnected network of data centers where you can really start to turn the workload of application in the clouds. When you start to compare costs, and better service performance workloads moving, it really starts to look like electricity.

Cole: So how do we go from here in there? What kind of investments should be looking for companies to move forward?

Sorofman: There are a series investments to be made to get there. The first is virtualization infrastructure, then the demand side. Level two is testing – Using Amazon's Elastic Compute Cloud as a basis for experimentation. Get your hands wet, dirty feet. Deploy some applications. Understand what the cloud looks like, and adopt some best practices to use as a basis for learning.

Step Next is to lay the bases of clouds. Take a reference architecture and codified as a set of processes and best practices that allow you to evolve. Then start releasing applications in the cloud on a limited scale, but enough that you begin to understand the complexity. Once you have architecture in place, the next step is to connect a circuit applications to the cloud, whether external or a cloud internal and begin the broad deployment of applications.

The last step is the achievement where you have this notion intergalactic the workload provisioning through the clouds. This is where you finally start to see real cost savings and Service features of your investment. But know that nothing is simple about it. It will be quite complicated, and many technologies are not yet mature enough to support this vision.

Read Art’s article, “The Cloud at the Starting Gate” – http://www.itbusinessedge.com/blogs/dcc/?p=550&nr=inbound

Just some simple thoughts