There are a variety of ways to relocate a data center involving a range of costs, downtime and stress. The one thing they all have in common, writes InfoWorld's Paul Venezia, is the need for lots of advance planning.

The easiest way to move a data center is to build a new one from scratch, and then install high-bandwidth connections to the old one and migrate virtual machines using virtualization tools. While this may be the easiest way and it minimizes downtime, it requires a fully virtualized infrastructure and it is very expensive, Venezia writes.   

A less expensive option is to build part of a new data center--racks and core networking gear, for example--before the move and then shut down service when you want to relocate servers and storage to the new site. This option takes a minimum of one day downtime, and it also raises the risk of lost data and services.  "It's also performed under the gun, as critical services and applications are down for the duration, and when a storage array stubbornly refuses to come up properly, that downtime grows as the new problem is dealt with," he warns.

The least expensive option is to move everything--racks, servers, storage, networking gear--from the old center to the new one. This takes the longest time and can be very complicated. Since the first method is far too expensive for most organizations, a combination of the second and third methods is the most common option.

Making sure it goes as smoothly as possible takes a lot of planning, beginning with anticipating what could go wrong.  Here are a few additional tips: 

• Be sure there is plenty of help available. "When the senior network admin is neck-deep in switching and routing reconfigurations, you don't want him distracted by quandaries like how the hell that blade chassis rail kit goes back together," Venezia writes.
• Map out detailed plans for where equipment--"down to the rack-unit level"--will be situated in the new center.
• Have a lot of photos taken of the old and new data centers during the migration, and make sure every piece of equipment is marked. "Label everything: the servers, the switches, the KVM dongles, and all the rail kits as they come out of the racks. Few things are more frustrating than spending an hour searching for the other rail for a mission-critical database server while the project comes to a standstill."
• Make sure good drivers are hired to transport the gear. "The hardware and data contained within those vehicles is more important than anything else in the company, and when it's rolling down the highway at 70 mph, the risks of losing some or all of it are extremely high. Putting an eager intern behind the wheel of a rented truck might be a bad idea."

For more:
- see Paul Venezia's article at InfoWorld

Related Articles:
How to squeeze more life out of the data center
Taking data centers to 100 percent virtualization