August 2009 Archives

Author: Lucian Lipinksy De Orlov, VIRTERA

Certainly a lot of press has been made of the potential impact of the H1N1 virus.  

According to the World Health Organization, "Many countries could see swine flu cases double every three to four days for several months until peak transmission is reached, once cold weather returns to the northern hemisphere.  Winter months tend to see increases in diseases such as influenza as people tend to stay indoors more and are in closer proximity to one another.  Southern hemisphere countries such as Brazil, Chile, Peru, and Australia have been experiencing dramatic increases in infections and deaths."

To make matters worse, the CDC recently reported that there is a shortage of school nurses.  The CDC recommends a workload of 1 nurse per 750 students, but an analysis puts the national average at 1 per 971 students. Why is this important?

With Windows XP approaching end-of-life and Windows 7 just around the bend, one thing is for sure: over the next few years the desktop is going to be different. Advancements in server-based computing have customers revisiting the idea of centralized desktop and application management. That's right - you heard me - we are going down that road again. I know you are going to say, "We already tried to do this years ago with Citrix and it worked great for some things but now I mainly just use them for remote access."

I too was part of that campaign to stop managing operating systems and applications at the endpoint and move them into the datacenter. When working for a previous employer, we designed and implemented a solid server-based application delivery system. We had fantastic meetings to congratulate ourselves on our brilliance and our new solution. 

This isn't another product pitch, nor is it another plea for needing an elaborate tool to track virtual machine usage metrics and calculate how much you should charge a business unit for X usage of a system.  Instead, my purpose here is to discuss exactly why you need to consider a chargeback strategy early when deploying virtual infrastructure.

Virtual infrastructure, whether deployed on an old standby like VMware or on an up-and-coming platform like Hyper-V, has costs associated with it just like all IT assets.  In most organizations, business units that require server infrastructure to support an application will typically pay for that hardware as part of the project.  The Business Owner will request a new system from IT and supply the software vendor's recommended hardware requirements for the system.  IT will then procure the hardware required for the system and charge the business unit for the assets as part of the project.

So, you've already virtualized all the easy stuff. You've replaced that rusty old hardware off of your datacenter floor and migrated legacy applications to a virtual infrastructure. You've even toyed with creating a disaster recovery plan that relies heavily on virtualization offerings from VMware and Microsoft. But that's so 2006.