|
Virtual sprawl, or the proliferation of virtual servers, is a concern because it's so easy to create and deploy new servers. As a result, organizations seeking to reduce complexity and costs by consolidating their infrastructure through virtualization could end up facing a different version of the same basic management issues they were attempting to resolve.
While purchasing departments may have their share of detractors, they have traditionally served a useful function by monitoring the proliferation of servers within a business. Virtualization means that IT organizations need to establish a new way to impose discipline and control server creation. The most practical approach is to define and enforce policies that require justification, parameters, and documentation around creating each virtual machine. You can also limit access and the ability to create virtual servers, and implement data center audit tools to manage virtual machines.
In most cases, production environments aren't a problem, because the people creating virtual machines are also responsible for managing them. This direct relationship means they're less likely to randomly create new virtual machines and thereby increase complexity. On the other hand, application development (AD) teams often don't have that motivation of management responsibility. AD teams might want to create new virtual servers for testing and staging, but afterwards they'll be underutilized and eventually abandoned – just as physical servers have traditionally been. If this becomes an established practice, IT departments teams will be faced with a real cost in terms of increased complexity and infrastructure management time.
|
