Posted by Reint Jan Holterman around lunchtime: June 28th, 2010
Virtualization and cloud computing have exploded over the past few years. A recent study shows that 90 percent of businesses have implemented server virtualization at some level and Gartner estimates the current market for cloud services is $46.4 billion and will reach $150.1 billion by 2013. With other estimates saying business applications can be made three to five times less expensive and consumer applications five to ten times cheaper, it’s no wonder it’s taking off.
Among other benefits, virtualization and cloud computing are helping companies remove physical dependencies from network resources, respond faster to changing IT-infrastructure needs, and lower overall costs. While this is all well and good, as companies add more and more virtual resources to their network, they struggle to keep track of them. The reason? Virtual sprawl.
Virtual sprawl is another word for the explosion of servers caused by virtualization. Every application and its middleware run on a dedicated virtual server, so instead of running 100 physical servers, an enterprise now may have 1,000 virtual servers. Robert van Loghem in his blog already explained this in more detail. And as with other growth problems, you may not be able to stop sprawl, but you can manage and contain it.
How does one manage virtual sprawl for deployment processes? Well, certainly not manually. Manually managing a virtual infrastructure is an impossible task. Virtual sprawl is all due to the uncontrolled creation and ongoing lifecycle of virtual machines. Virtualization is inherently dynamic and flexible, allowing companies to create, copy, change and move things much more quickly than ever before. However, the ability to track and manage these virtual assets has not kept up. So, when an actual task is at hand, like deploying the development team’s latest application, it’s impossible to navigate all the virtual blind spots while tracking down and connecting the right servers and middleware components.
Deployment automation and middleware automation is, therefore, a necessity.
Modern-days deployment automation solutions have built-in support to discover remote servers, whether they are in a private or public cloud, and make them available for application deployment. When discovering remote servers, these solutions discover what middleware is running and how it is configured on a company’s servers. They can also discover clusters of servers for deployment within mere minutes, allowing companies to quickly scale up capacity when needed.
Deployit 1.3 beta edition, just announced earlier this month, is an example of one such solution that can consistently manage the explosion of virtual servers and rein in issues of virtual sprawl through its repeatable automation process. So, when deploying applications in virtual and cloud environments, it’s important to avoid the frustration and errors associated with manual processes and virtual sprawl by first turning to deployment automation solutions.
