Categorized | Business

Preventing, managing datacentre emergencies

Even an expertly engineered and thoroughly commissioned datacentre cannot guarantee 100% availability. Datacentre operations and maintenance teams should always be prepared to act swiftly and surely, without warning, during emergencies.

“Good preparation is the best defence and will help ensure responses are timely, effective and error-free,” says Riaan de Leeuw, VP for IT Division – Anglophone Africa. “Schneider Electric has drawn up a 15-page White Paper, ‘How to Prepare and Respond to Data Center Emergencies’, which we know will be beneficial to all datacentre operators.”

Table 1 below gives a short overview of key aspects of an effective emergency preparedness and response programme for data centres. There are seven key elements, which are grouped within three higher-level categories.

The paper describes each element and offers practical advice to assist in implementing this strategy.

Be prepared
Preparedness begins with developing emergency operating procedures (EOPs) for all identified high-risk failure scenarios, such as the loss of a chiller plant, failure of the generator to start etc.

“Of course, what constitutes a ‘crisis’ or ‘disaster’ may vary from organisation to organisation, depending on what they deem to be critical and urgent. A Tier IV data centre might have a different definition than a Tier I site, with site and software redundancy in place,” adds de Leeuw.

“The best crisis management tool is prevention and here the crisis management plan (CMP) is essential, with its set of policies and procedures to help data centre operators prepare for, respond to and learn from crisis. The plan is designed to instruct people on how to detect/prevent and react to a variety of crisis scenarios, with the goal of providing a safe, timely and sound resolution that prevents the crisis from evolving into a total disaster.

“Emergency drills should be mandatory and should be created for each EOP that addresses anticipated events of high probability and/or high severity. Each facility should establish a goal of each data centre operations team member participating in at least one drill per month but in all circumstances must meet any contractual obligations regarding drill requirements.

“Finally, incident reporting should be designed so that it focuses on Why the situation occurred rather than the Who, What, How, When and Where.”

“Companies that struggle with adequate response to natural and man-made risks should seek the assistance of critical facility operations subject matter experts. By implementing the best practices developed over many years in Schneider Electric, organisations can protect their expensive assets such as datacentres and ensure the best return of investment,” concludes de Leeuw.

Share this article on Social Networks facebooktwitterlinkedinmail

Leave a Reply

Your email address will not be published. Required fields are marked *

TransformSA - Subscribe

Subscribe - Transform SA Magazine

Transform SA in conversation with Nompendulo Mkhatshwa about student movement.

Photos on flickr