Server virtualisation has revolutionised the way businesses store and access information, saving companies millions in hardware and infrastructure costs. When applied to disaster recovery (DR) and backup,
Virtual disaster recovery (VDR) is a relatively new concept that can get all systems back up and running quickly and efficiently without the need for expensive additional hardware or remote sites. It's the next natural step in the evolution of disaster recovery (DR), and when outsourced to a third party provider, it can be delivered as a completely managed service.
VDR is the logical step between companies managing their own IT and outsourcing it to more cost effective service providers.
Peter Groucutt, Managing Director of Databarracks,
Virtual Disaster Recovery: What it is and how it
Traditional disaster recovery was incredibly expensive, as everything from a data centre (which could be as simple as servers in a room) had to be duplicated to a second location. Virtual disaster recovery means that both of these data centres can operate running constantly and simultaneously.
The process of configuring, deploying and testing Virtual Disaster Recovery remains the same. Databarracks is operating VDR right now, but every business has different requirements. The applications in place and how they are used can change dramatically, so a lot of work is done to ensure that the recovery system is tested.
In the event that server failure occurs and major DR is required, Databarracks finds and recalls the latest virtual server snapshot. The image of the server that suffered the data loss is then repopulated without having to reinstall or reconfigure any aspect of the virtual machine. Because the restore is incremental, rebuilding the most recent server image is a short process.
Traditionally, businesses would implement a DR system based on its low price instead of the convenience and ease of use it offers. By contradiction, most companies also demand zero server downtime to ensure business-critical computing is maintained. Virtualised disaster recovery ensures that downtime is dramatically decreased for a fraction of the price of duplicating an entire data centre.
Best practices for implementing a Virtual Disaster Recovery plan
After considering what information and applications a business needs, a DR plan can be designed. A business will have more data than is actually required for the business to survive in the first few hours of a disaster. All documents will be needed eventually, but only the active documents are needed immediately. An effective DR plan will balance the needs of the business with the technical practicality of bringing information and applications back online.
Databarracks uses NetApp products to power its VDR processes. The company uses a combination of FAS3040 filer heads and 6000 series. Depending on the type of environment that is recovered, Databarracks deploys SATA and SAS disk configurations. For file data, SATA is used. For Exchange information, databases are used and for demanding applications, SAS would be the underlying storage.
Managed online backup and outsourced Virtual Disaster Recovery
Many companies are now choosing to outsource their backup and Virtual Disaster Recovery to a specialist backup provider. These providers will manage the cost for less than the company could do itself, as well as offering both access to the latest technology and the manpower to manage a DR project quickly with low impact on internal IT resources.
If companies choose to manage their own virtual DR strategies, their investment will include hardware, software and hosting capabilities. This is expensive for a platform that is only used occasionally. A managed serviced provider will keep hardware available for a company to use when necessary, substantially lowering the cost. Most managed backup providers also offer Virtual Disaster Recovery as well as a one-stop shop for same day backups and recovery.
Outsourced Virtual Disaster Recovery is a real option for businesses that don't have the time or budget to manage DR internally, and as awareness of this service grows, we can expect to see more businesses look to outsource the entire DR function.
In the last few years, businesses have started to realise the full advantages that virtualisation offers to their production IT environments. In the next few years, they will also be able to accelerate the advantage of virtualisation through cloud computing. Companies no longer need to spend large amounts of capex on disaster recovery; production environments can now be made available quickly and affordably without the risk of wasted investment on hardware as an insurance policy.
The more companies start to become comfortable with utilising hosted and managed IT systems in the cloud for their disaster recovery strategy, the sooner they will become willing to look at placing their production IT infrastructures there, too. Virtual Disaster Recovery is the logical step between companies managing their own IT and outsourcing it to more cost effective service providers.
Peter Groucutt, is the managing director at vendor independent online backup services company Databarracks and a Contributor to SearchVirtualDataCentre.co.uk.
This was first published in January 2010