What makes for an effective disaster recovery plan for a data centre?

It is quite an irony if you look at it. Data centres exist to keep your data safe, guarded and protected; however, even DCs need a disaster recovery plan. Any of the natural calamities like floods, fires, earthquakes etc. can render a data centre powerless or worse, destruct it irretrievably. And, these are not the only kinds of disasters that a business needs to prepare for; disasters can be caused by unexpected power outages, viruses attacks, or even an employee that accidentally deletes sensitive files, hence redundancy of data cannot be stressed enough. Besides data loss, downtime can cost business significant amount of money; usually in the range of thousands of dollars per minute.

It is therefore extremely pertinent for data centres to have a Disaster Recovery (DR) plan in place and that too an effective one. A DR plan focusses on data protection in most cases and also on keeping the information stored as recoverable as possible. However, an effective DR plan must also focus on equipment protection along with the estimation of power demands in worst-case scenarios, cooling requirements and other security considerations.

Here are some of the factors considered in a DR plan:

Risk Assessment

The first step is to access risk! List out the crucial systems that keep the business running and noting damages that could occur in the event of a disaster. This list should detail the risks for every possible disaster the business is susceptible to, as well as the costs associated with the downtime suffered. The list should be comprehensive and should flesh out all possible scenarios. Once this is done step-by-step plan can be created to deal with each risk.

Location Selection

To begin with, data centres need to be planned to be built in areas that are not prone to natural calamities. That would be ideal and also quite obvious. However, Mother Nature is unpredictable and hence it is also recommended that the equipment like servers and other storage devices are stored in a facility that is compliant with global standards and has relevant certifications and accreditations.

Data Redundancy

Data redundancy cannot be emphasized enough. Be it the cloud or varied data centre facilities, back up is more relevant than ever. Finding, a good secondary site in another geographic zone is the foundation of a DR plan. Apart from a backup of data, it is required to create redundancy for most things like power back up, cooling back up etc; in short, a fully redundant, mirrored system.

Periodic Risk Assessment Tests

Every data centre must be ready with a DR step-by-step checklist and run periodic tests and drills to check if the measures taken are effective in an adversity. Also, every employee involved must be thoroughly trained in carrying out DR processes, otherwise the plan stands beaten.

Look Beyond the Data Centre

It becomes relevant to gauge the local infrastructure around the suggested location of the data centre too. If the generators run out of fuel, is there a provision in the vicinity. Also, there shouldn’t be reasons around that create moisture in the data centre system making it more prone to an outage. If your wiring is kept underground in order to keep it safe, what are the chances that the water table may rise due to the surrounding activities? How soon can help arrive in case of an adversity? Are there multiple ways of getting there?

Predict the Changes

With the advent of global warming, there are now many studies available that may help you understand the future risks better. It is important to choose your cables wisely, keep them moisture-proof and use technology to optimize them to any imminent danger. Sea levels are increasing, coastal areas are getting flooded, there is a surge in storms etc. Your infrastructure might be water -resistant but is it capable of being submerged in water if such a situation arises? For every degree of rise in temperature, the data centre must be capable of withstanding the worst of weathers in all extremes.

At STT GDC India, we understand the possible impact of natural disasters significantly and thus our sites are planned with utmost care. STT Hyderabad DC1 is one such site that has been built at one of the least calamity prone regions of the country. And yet, we have made sure to create N+1 redundancy with 24 hours on-site fuel storage capacity for the transformer and DG and a distributed redundant UPC system with a backup of 13 minutes at full load. Also, the power sources to the UPS originate from two independent PDUs. It is ISO 14001 certified for being sustainably designed and is TIA -942 compliant.