Disaster-proof virtualisation on a dime: how I did it

Most companies virtualize servers to save money, save space and save time for the business. HR outsourcing firm The Sullivan Group had another motivator: hurricanes. Here's their thrifty, successful formula for virtualization, using Citrix and Marathon Technologies products.

"Marathon has been working with Citrix for a long time, and they have the ability to control host-based replication and extend that over distances, so that replication could really be the driver for a company to deploy if they're looking for HA," Bowker says. "We already know Windows Server 2003 runs on a lot of VMs, so there's no reason XP or Vista can't do the same thing.

"[Sullivan] doesn't have a VDI infrastructure in place, but even that's not too unusual in that kind of environment," he says.

The reality that there are many products from many vendors that accomplish similar things and can be fit together to accomplish a specific set of goals tends to get lost in most virtualization marketing, Bowker says.

"A lot of my conversations with vendors are about that," he says. "Everyone has their own agenda and the way they have to slice things, and reality shouldn't intrude. There's a lot more going on in most data centers than people think."

Going off the main path just requires a little extra knowledge, or a little extra work, Wolf agrees.

"It wasn't that complicated to do," Simpson says. "We were up in about a week and haven't run into any big problems."

Well, there was the RAID card that failed in one of the new servers a week after the upgrade; but as the server failed, another took over and Jones and Simpson found out about it later.

"We didn't even know it happened until it was done," Jones says.

Follow everything from CIO.com on Twitter @CIOonline

Tags disaster recoveryvirtualisation

Show Comments