Setting up a cold site for disaster recovery can be a great solution for organizations that don't have a stringent recovery time objective (RTOs). Harvey Betan, associate principal at Risk Masters Inc., defines what a cold site is, the pros and cons of cold sites, and alternatives to cold site disaster recovery in this Q & A. Learn about what you can expect to get when you lease a cold site vs. a warm or hot site, what organizations should consider cold sites, and other alternatives to cold site disaster recovery. You can read his answers below or download an MP3 of the conversation.
How do you define a cold site for disaster recovery?
A cold site is space you have reserved in another location, and you can you do what you want with it—it’s empty space. So, you have to configure the space as you want for people to sit there, set up hardware, servers, PCs, etc. It’s basically an empty space for you to use if there’s a disaster on your site.
What do you see as the main value for a cold site? As opposed to setting up what we’d call a warm site or a hot site?
Well, a cold site—since you have no hardware there and it’s an empty space—is a lot cheaper; you’re just reserving the space. So that’s from the financial aspect. If your systems are down or for some reason you can’t operate on your normal environment, you’d want to have another location to be able to get your system up, be able to have access or your staff have access it so they can continue to work.
So, who is this going to work for? Who would be interested in this approach? What kind of organization, what kind of needs?
It would work best for somebody who doesn’t have an RTO, which is setting up a technical environment quickly. And anybody who is in a small business that is not transaction-based, because you’re not paying for the hardware to be in place.
It is not good for someone who is transaction-based, because you are losing more time to configure the space and then start setting up your equipment, and it’s sort of a dead time.
Can you talk a little bit about what is a warm site and what is a hot site, and what are some of the pros and drawbacks of each route?
The reason you’d chose one over the other is basically what your requirements on the technical side are for getting your systems up. With a warm site, you have the equipment there, but you have to restore your data… probably not the operating system because that’s already established. So your equipment is there, but not your data.
The hot site basically means your equipment there; it’s running, your data is pretty current so you basically flip the switch from your primary site to your recovery site and off you go. Obviously, that’s the one that is most expensive.
What about other alternatives? For example, there are more and more cloud services becoming available. Is that really an option—to set up a DR site in the cloud?
Yes, there are areas that are using DR sites in the cloud. But then what you have to consider is not only are you outsourcing your processing to someone else—people use the term “cloud,” but it’s not just sitting up there in the sky, there is an endpoint—the hardware actually works. But now your single point of failure is your communication. You must be able to communicate with the cloud to be able to process your data. You have to start worrying about the speed of transmission, how much data you’re sending back and forth, those issues come about, but that’s for another conversation.
This was first published in May 2011