What does IT disaster recovery (DR) mean to you? Does it bring to mind racks of equipment sitting in a computer centre somewhere powered down or in an idle state waiting to be cycled up? If this is the case, it might be time for you to think again.
Modern businesses rely on IT in order to function efficiently and effectively. When single applications go down, the impact on business operations can be dramatic, and not only for so called 'mission critical' systems. If the disaster effects not just a single application but multiple systems or your computer room or data centre, the impacts can be severe. Yet DR in many companies is confined to a small number of mission critical applications.
So, if it has been a while since you thought about your DR measures, or a review has been prompted by a risk assessment, compliance audit, actual disaster or some other scare, it's worth taking some time to understand what can be achieved today.
A new paper by Freeform Dynamics, 'The Democratization of IT Disaster Recovery' discusses how DR capabilities enabled by developments in the IT landscape that have taken place over the past few years may well make it possible for you to broaden the scope of systems you could protect. In particular growing maturity in four areas are together providing new DR options, especially to mid-size and smaller organisations.
The first area concerns advances in wide area networks and systems to optimize WAN connectivity. In the context of DR, this means you may not have to worry so much about providing an alternative physical place of work should an office or building be rendered inaccessible. This opens the door to data protection strategies that back up or replicate data across the network, whether site-to-site, between home offices and the data centre, or between company and third party facilities to provide DR for computer room systems.
Such capabilities could also make it possible for you to better protect against disasters with desktops, laptops, tablets and smartphones. This rarely grabs attention, but with many employees reliant on such devices to do their jobs, improving recovery times from problems, which happen frequently, can bring significant productivity advantages.
The second significant development is the growing adoption of infrastructure, especially server, virtualisation. Server virtualisation essentially means a server and its software stack is converted from a physical asset to a data object which can be moved around.
In a DR context, this allows servers and applications to be replicated to any location and spun up on alternative hardware should you need to recover it following disaster or failure. This opens up the potential that DR capabilities may no longer require having identical physical servers waiting around 'just in case' to get the business back up and running.
With suitable software, many VMs can be recovered to any hardware that presents a compatible hypervisor, on another system in the business or in 'the cloud'. There's a little more to it, but details can be found in the paper referenced above.
Speaking of Cloud, the third area of advancement concerns the growing maturity of cloud services, whether from large players or local service providers. One of the big selling points of cloud is the ability to acquire resources rapidly, even 'on demand'. Depending on the arrangement with your provider, the compute and storage capacity needed to recover the servers and data associated with a business application can be available within minutes or hours. Some cloud-based DR options may also have the benefit that you only start paying for resources when you use them.
Cleary there are elements to consider, including how to keep cloud-based servers and data in sync with operational systems, and/or how to get them to the target cloud environment in the event of an emergency. Naturally any use of cloud services must take account of security and compliance requirements as well as how to 'transport' data securely and cost-effectively. This could be something where the WAN optimisation technologies mentioned above may play.
The final topic to consider concerns the considerable advances made in management solutions, including the often overlooked area of backup and recovery. Modern solutions are available now to replicate or synchronise data between environments, even over extended distances. Such operations are pretty straightforward on slow moving systems, but it's likely your business is dependent on one or more applications with high rates of data change. Historically these have been more difficult to provide with effective DR, but modern protection solutions are able to cope with data stores that change rapidly.
In effect it is now possible in many scenarios to operate data protection solutions which minimise the loss of transaction data in the event of failure without inflicting heavy overhead on live systems. More sophisticated solutions are also able to recover complex systems composed of several VMs, without the need for complex manual scripting.
The advances in data protection management solutions, virtualisation, networks and cloud systems make it possible for cost effective DR capabilities to be provided to an expanding range of your business systems. If your DR is limited to just a few 'mission critical' platforms you should find time to work out if you could cover a lot more of the applications which run the business.
Simple first steps you should take include taking an audit of the applications your users employ every day. Working out which do not already have DR plans in place, and of these which should have better DR coverage. It's then a matter for you and your teams to decide how to provide the DR they require.