Cloud computing‘s popularity is on the rise. Service providers like Amazon are providing public cloud hosting at rapidly decreasing prices, while bandwidth capacities are increasing. However, just because it’s virtual doesn’t mean it’s infallible.
Cloud computing outages – from a failed component to total service meltdown – are inevitable. During such events, businesses will lose access to data and resources and suffer downtime, impacting revenues. Recovering from these disasters depends upon the strength of the business’ cloud outage contingency plan.
The importance of a contingency/business-continuity plan cannot be emphasized enough. You already have one for your physical network, so build a cloud computing plan that covers all eventualities — or as many as possible.
A good cloud outage plan should include contingencies for data recovery, continuity, and (in extreme cases) infrastructure failures. The plan’s parameters will vary, depending on the size of the organization and data usage patterns. Here are 4 tips on where to start:
Know thy cloud computing network
Keep updated lists of dependencies, data access permissions and data controls in your contingency plan. If you don’t know what’s running in your cloud, you’ll have a hard time trying to put things back together.
If your business is heavily dependent on the cloud for key services and mission-critical data, you need to create a comprehensive map containing core components and the order of recovery. In most cases, an organization’s entire data need not be recovered to continue functioning during a cloud outage. However, knowing what to recover and the correct order, can save time and effort in case of such an event.
Address user experience issues
The end user will probably feel the full effect of a cloud computing outage, so factor user experience into any contingency and/or backup plan. Aim for minimal disruption from a user’s point of view during the planning and design phases. A properly formulated strategy will allow the user access to alternative resources in a specified time, usually according to a service-level agreement.
Master cloud traffic control
Controlling wide-area network (WAN) traffic in the cloud results in optimal network performance, and helps direct users in case of a networking-related cloud outage. In a cloud environment, data may be spread across multiple locations and access times, and when added up may considerably slow the process. It is good practice to understand WAN traffic and optimize the network by creating highly efficient network paths between data sites. A number of software solutions, such as Silver Peak and Riverbed, handle this traffic easily and can be deployed at the virtual routing layer.
It is imperative for any environment to be tested extensively, with subsequent documentation. No matter how foolproof a contingency plan looks, it needs to be tested in real-world working conditions to ensure dependability – especially if an organization is based within cloud infrastructures. Maintaining proper documentation on any tests can prove invaluable in a similar, actual event.
The exact cloud outage contingency measures vary from business to business. However, if you know exactly what is running in the cloud, what your users’ data needs are, and what data is essential for your business to continue running smoothly, planning for disaster should not be too much of a hassle.Tags: Business Management,Cloud Computing,Productivity,Technology,Virtualization