How well can your organization deal with an emergency? Automatically sign up for our free Disaster Recovery newsletter, delivered each Tuesday, and make sure you're prepared for the next catastrophe.When dealing with disaster recovery and business continuity planning, organizations sometimes focus more on the nature of the disaster and less on the actual solutions needed to protect the company's operations. After all, discussing potential earthquakes, floods, or even a power outage is usually more exciting than weighing the merits of tape backups over replication tools.
However, the solution your organization chooses can really make a difference. You have a myriad of recovery solutions to choose from as well as a large selection of vendors.
At the end of the day, nearly all solutions fall into one of three categories of recovery protection. In many cases, the same product will address more than one type of recovery, so it's important that companies do their research to find the best combination of price and functionality flexibility.
To start you off on the right foot, let's review these three major categories. No matter which combination of technologies you use, knowing the differences between them will help your organization find the right solutions for its business needs.
While often used as a catch-all term for any recovery method, the definition of DR usually entails the restoration of data from a copy to a system that's either available or quickly acquired and configured for use. The basic concept behind this term is getting data off live production systems and keeping it someplace safe.
The most common example of a DR solution is a tape backup system, which allows you to store a copy of any data on a server system in any location that you'd like. This definition can also include systems designed to store data on spinning disk, instead of offline tape, either locally or in a remote location.
The main benefits of DR systems are that they're generally cross-platform and centrally managed. This makes them very easy to use under most circumstances; however, as with any technology, you can easily find yourself in over your head if you don't work things step-by-step.
The main drawbacks of DR solutions are that you need to move tape-based media off-site, or you risk losing the taped data along with the production systems in a disaster. In addition, you must restore the data to another system before you can resume operations, which could take a significant amount of time.
HA solutions refer primarily to systems that offer the ability to immediately resume operations with a minimum of downtime using new hardware that you've preconfigured at the same physical site to stand in for a failed machine. While it's theoretically possible to provide HA to another physical site, these solutions often demand the same networking properties (in terms of IP address and other settings) that the production system uses.
Most systems combine some form of replication tool with monitoring and failover automation tools to provide this functionality. In some cases, you must use a combination of vendor tools to get both sides of that equation.
The main benefits of HA systems are that they offer the ability to resume operations without restoration efforts, typically with very little downtime, and little to no client-side reconfiguration.
The main drawbacks of HA systems are that they tend to be more expensive than simple DR solutions, and they also tend to be platform-specific due to the complexities of the failover process.
RA solutions take the idea of HA and extend it to include other physical locations and networks. While the same principles apply, you now need to deal with WAN links and other limiting factors that don't usually exist within a single data center.
You must also determine if end users will be able to work with the systems in another location. In addition, you'll need to update client-side apps to make them function properly.
For all of these drawbacks, the benefits to RA systems are simple and profound: You gain the ability to survive a disaster that incapacitates an entire data center. This level of insurance can easily become well worth the expense and headaches of implementing it.