7 half-truths about virtualization
Network World - Virtualization is revolutionizing the data center, mostly for the better. But no technology is without potential pitfalls. Problems related to management, security, ROI and power use can all trip up a virtualization deployment that isn't planned properly.
"Virtualization has the potential to deliver immense cost savings and technical benefits," through the consolidation of servers and reduction of space and power needs, notes Laura DiDio, lead analyst with Information Technology Intelligence. "However, these savings don't automatically happen."
Here are seven -- oh, let's call them half-truths -- to fully consider before your virtualization project is implemented.
1. Virtualization will make my life easier.
Virtualizing servers will greatly reduce the time it takes to spin up new workloads. Some IT shops have reported being able to deploy new virtual machines (VM) in as little as 30 minutes, as opposed weeks for their physical counterparts. The promise that virtualization simplifies IT is real in many respects.
But virtualization simultaneously introduces management challenges that can't be ignored. IT shops need strict policies and perhaps third-party automation tools to prevent virtual server sprawl, the unchecked spread of VMs. Even if you end up with fewer physical servers, Burton Group analyst Chris Wolf says the overall number of managed objects can increase, because of the hypervisors and sheer number of VMs.
Many users assume administrative time will be lessened, but in reality the virtual infrastructure itself has to be managed and may require a new centralized storage system, says Martijn Lohmeijer, managing consultant with TriNext, an IT outsourcing and consulting firm.
Frustratingly, many software vendors don't offer the same levels of support for applications running in VMs as they do for applications running on bare metal. Microsoft eased up on some licensing restrictions last year, but analysts are still criticizing Oracle and other vendors for restrictive policies related to support in virtualized environments. The calculation of software licensing fees can also be more complicated in a virtual data center.
"Not all server virtualization licenses are the same," DiDio says. "You have to really study the terms and conditions of your licensing contracts from the various vendors."
2. Consolidating onto fewer servers will be simple
The first goal of a server virtualization project is often consolidation. If you can run the same number of workloads on 10 servers that you were running on 100, why not consolidate as fast as possible? Unfortunately, many IT shops that plan to consolidate end up doing so much slower than they expect, says George Pradel, director of strategic alliances for virtualization management vendor Vizioncore. It's easy to say "every new workload has to go on a virtual machine," but moving old workloads from a physical box to a virtual one is not always a simple task, he says.
Reprinted with permission from
Story copyright 2009 Network World, Inc. All rights reserved.
Your virtualization deployment could be tripped up if you don't consider these potential roadblocks.
Additional Resources



Learn the important issues you must consider before starting your next mobility initiative. Get your mobility white paper from IDC now, compliments of Sybase.
White Papers & Webcasts
IDC Research Report: The Business Value of Consolidating on Energy-Efficient Servers
Download this Resource Now!
The Business Case for Virtualization
Download this Resource Now!
Low Administration ROI Tool
Download Now
Clipper Group Report: HP Provides Enhanced Options for Data Center
Download this Resource Now!
Effectively Implementing Datacenter Automation
Effectively select and deploy the best datacenter automation solution today!
XenApp Extends Virtualized Application Delivery
Download this webcast to learn how to accelerate delivery of virtualized applications and streamline management.
Top HPC Use Cases in Life Sciences
Learn from the experts how best to apply cutting edge high-performance computing techniques a life sciences environment.

