By Dan Barber
Many companies don’t do Disaster Recovery because they think they can’t afford it. It’s true that setting up a DR site can be expensive, but there are ways to keep the cost down. And the number one way is to deploy used or repurposed storage.
In most scenarios, your DR site isn’t your primary IT environment. It’s like insurance—you only need it when you need it. That’s why you don’t have to buy a top-of-the-line storage system; it’s not going to be used on the same level as your production SAN. Instead, a DR site is a great opportunity to repurpose older storage and/or buy used storage to mirror part of your primary environment.
1. Which users will be logging in after a disaster?
In planning your DR site, you should ask “What percentage of our users will be logging in after a disaster? And “What level of performance is acceptable from a business perspective for that group of users?” The answers will help you determine the storage capacity and level of performance you need from your DR site. For instance, if a hurricane were to destroy your city, do you think all your users will be logging in? Not likely. So while you may still want performance at your DR site to be similar to your production site, you don’t need to size the site to accommodate all your users. Preparing for a much smaller set of users can reduce the cost of your DR site substantially.
2. How is used hardware advantageous?
Used hardware is also a cost-saver because you can skip worrying about firmware and software upgrades. You want your site to be up and working, but don’t want it to consume a lot of your time. If you have to go from managing one whole site to two whole sites, you’ve added a lot of soft costs. Instead, the goal is to set up the site and be able to pretty much forget about it, which you can do more easily with used hardware.
One DR option companies often consider is buying the exact same equipment they have in their production environment. But remember how costly that environment was to set up. Used storage can slash those costs dramatically. You can buy the same basic hardware in a similar configuration with less capacity, and let that handle DR. Administrative time and costs stay low because you’re already familiar with the system.
3. What are my options for replication to my DR site?
Regarding replication of data to your DR site, keep in mind that there are many methods to do that. The key is to choose the one that best suits your setup. You may use storage array replication where, for example, a production NetApp device talks to the NetApp device on the DR site and does replication on the backend. This kind of replication is independent of applications, and works with file servers, because they don’t generally have replication built in. But for Microsoft Exchange and SQL Server, Microsoft doesn’t recommend doing array-based replication; they recommend doing it within their application to make sure the data has consistency at a management level. They want Exchange, and not the underlying storage, to be in charge of copies. Many users also do replication through VMWare, which is even more prevalent than array-based replication.
Bottom line: A DR site is smart planning and good insurance against the kind of disaster that can seriously damage your business. Don’t avoid it for cost reasons; think used storage, and take the cost-effective route.