Nobody likes to think about it, but at some point, data centers will fail, either due to natural or man-made disasters. Having a disaster recovery plan in place can help mitigate service disruptions and the loss of data and revenue that can occur in the event of outages in production systems. In this article, we take a look at the key metrics in disaster recovery as well as what options are available to organizations.
Key Metrics: RPO and RTO
In the world of disaster recovery, there are two key metrics: Recovery Time Objective (RTO) and Recovery Point Objective (RPO). RTO refers to how long your business can tolerate being unavailable – in other words, how quickly do you need to be able to start working again? The lower the RTO, the more you should expect to pay for disaster recovery services. RPO refers to how much data your business can tolerate being lost in the recovery process, measured in anywhere from minutes to days. In essence, RPO tells you how often you’ll need to make new backups. As with RTO, the lower your RPO, the more expensive it will be to design and implement a disaster recovery plan.
Keep in mind these metrics are meant to be more guidelines than hard and fast rules. Actual recovery times and points will vary somewhat, depending on the nature and scope of the outage. The key here is that RTO and RPO, along with budgetary considerations, should help point you toward the kinds of disaster recovery options you should consider.
Cold and Warm Sites
The most basic goal of disaster recovery is to prevent your organization from having to start from scratch when a major outage knocks out your data center.
A cold site is the most basic disaster recovery option. Sometimes a cold site will involve regular deliveries of tape backups, but it can also be as simple as a room with HVAC, a power supply, and network access. In these cases, the organization is responsible for installing the new hardware necessary to get the data center up and running again. Recovery time in these cases is best measured in days or even weeks. The upside though is that cold sites are relatively cost efficient to maintain since you’re really only paying to rent the space until it’s used, though that doesn’t include the cost of replacing the physical infrastructure that may be required in the event of a total disaster.
A warm site, on the other hand, already contains much of the basic infrastructure needed to get a data center up and running. As with cold sites, the term encompasses a wide variety of options, from dedicated spaces with regular backups to shared resources across multiple data centers. Previously, tape-based backup was the norm, but now electronic vaulting is much more common. Whether or not a warm site receives regular backups, it’s likely to be both more expensive to maintain and quicker to boot than a cold site, with RTO times ranging from minutes to hours depending on the exact setup.
The rise in cloud services has given organizations another option when it comes to recovery solutions. The ability for Disaster Recovery as a Service (DRaaS) providers to leverage the geographically dispersed gives cloud-based options many of the same advantages as other cloud services. For the end user, that means increased scalability, as well as the ability to continually backup their data and applications without worrying (as much) about the physical security of any single data center. On top of that, DRaaS options allow organizations to create highly redundant, mirrored data backups that run concurrently with their production servers. This can cut RTO and RPO to minutes or even seconds, which may be desired for organizations that require near constant availability.
Advantages of cloud-based disaster recovery:
- Geographically dispersed – No matter where a disaster strikes, there’s a good chance you’ll still be able to access your data and applications.
- Speed – Near instantaneous recovery times.
- Scalability – Quickly provision new resources without worrying about hardware.
- Elasticity – Quickly move resources between private and public clouds as needed.
Building a Disaster Recovery Plan
The first part of preparing for a disaster is creating a plan based on your organization’s specific needs and budget. (This is where RTO and RPO come in handy.) An IT expert can help you sort out the specifics.
Remember: Not every part of your organization may require the same level of availability. For example, archived customer data that’s stored for compliance purposes probably doesn’t need to be accessed as quickly as business-critical data. It may make sense both from a technical and a financial perspective to triage your data recovery plan so that resources are allocated toward getting your application and critical data back online first. In these cases, tape backup or cloud-based cold storage services may be an attractive and cost-effective option for storing important but not time-sensitive data.
This article was written by Tyler Keenan from Business2Community and was legally licensed through the NewsCred publisher network.