Here is an interesting bit of information about disaster recovery that is guaranteed to frighten any business owner, IT manager or systems integrator: Gartner studies have shown that 76% of companies experience an outage each year, and 40% of companies go out of business if they cannot access their data for 24 hours. Those are quite deadly numbers.
So what are the main causes of downtime?
- Hardware/software failures
- Human error
- Natural disasters
Naturally, you may experience a combination of all these factors. ING Romania’s recent 10 hour downtime comes to mind as a great example of human error combined with hardware failure. The drawbacks your business suffers in such a situation are clear: loss of money and, most importantly, credibility.
So what can you, the business owner, IT manager or systems integrator do in order to make sure your business does not suffer from downtime?
Choose the right infrastructure for your business
A solid IT infrastructure with integrated disaster recovery capabilities is vital. Various businesses naturally have different IT needs as far as capacity and functionality are concerned. Keeping this in mind, what should you take into consideration when choosing an IT infrastructure for your business? Here are a few tips:
Choose an infrastructure that fits your workload but is also capable of handling unforeseen events. Which applications are vital for your business? Do you need an extra degree of security for some data? For example, choosing an all-flash array for business-critical applications, like OLTP databases in banking, can keep your core applications running even if the server room is affected by severe vibrations (jet-engine-like sound waves generated by a faulty fire prevention system).
Your data infrastructure should be able to perform well with your constant workload. How many IOPS does your system need? Which applications should run at their highest performance? And finally, what type of disaster recovery matches your performance requirements in terms of RTO, RPO and resource overhead? If your datacenter is hungry for resources, a 3rd party backup solution which eats up IOPS and bandwidth might not be the best choice.
Last but not least, how much space do you need? When answering this question, keep in mind that new software-defined infrastructures use data-compression and deduplication algorithms that will “squeeze” the same amount of data in 2-3x less space. Also, a data replication system with zero-space allocation snapshots might be the perfect fit you.
Set up a disaster recovery plan and rehearse, rehearse, rehearse
A powerful disaster recovery solution does not equate to much if your DR plan is not executed to perfection during a disaster. With a wide range of natural disasters and human errors that could harm your business, you should always rehearse and plan ahead. Always test various scenarios in order to establish a solid disaster recovery plan for your operations and achieve the lowest RTO possible.
Here are a few tips on what to take into consideration when testing a disaster recovery strategy:
Test the most common scenarios
Having a foolproof DR plan is no easy task. Unfortunately, unexpected things may sometimes happen. But how can you react quickly in unexpected situations if you haven not tested out the most common disaster scenarios?
What are the most “likely” disasters that could interfere with your business? Start by identifying the most common scenarios and start your testing there. If your team is familiar with these scenarios, they will most likely react promptly should a real disaster occur. Ready for some next-level testing? Create an unlikely scenario to try out with your team. This will help you get creative and understand your needs and capabilities better.
Identify your main business needs
Not all businesses have the same needs. There is no “one-size-fits-all” approach when it comes to disaster recovery.
Which are the applications that are vital for your business? Maybe you need access to your databases or maybe you have one application that makes sure your customers are happy. Whatever it is, make sure you give it the proper importance and set it as a priority in your DR plan.
Analyse the interdependencies of applications
Each business operates using various applications that are more or less interconnected. The intricacy of applications is getting more and more complex. How does the malfunction of one application affect the rest of them? Testing a disaster recovery scenario should look into these connections.
Constantly evaluate business impact
Creating a DR plan once may not be enough, if your business is a dynamic one. How will a disaster affect your business? Today’s answer might be different than a few years back. Make sure you are aware of the business and IT impact in order to prioritise business needs and adapt your plan.
Train your team for a disaster
In order to make sure you are responsive in case of disaster, your team must have access to the same level of information. Set up a plan and procedures (line of communication, tasks etc.) covering all scenarios. And of course, keep testing.
All of these elements are to be considered if you want to make sure your business keeps going in case of a disaster. The right infrastructure, a tailored DR plan and a well-rehearsed strategy can keep your business going in case of a disaster, with minimal losses.
Click the “Ask us anything” button below if you have any question and we will gladly help you out!