Digital Realty is a world leader in data center uptime and reliability due in no small part to our preparation and planning around the handling of critical incidents in the data center. (And to learn more about how we avoid having critical events in the first place – it should be noted that data centers owned by Digital Realty experienced zero downtime following Hurricane Sandy! – check out this blog.)
Recently, I spoke with Joe Reele, Solutions Architect VP at Schneider Electric, about how Digital Realty responds to data center events, and how an effective response can lead to future success.
In our conversation, “Effective Response to Crisis Management,” I talk about Digital Realty’s approach to managing critical incidents, which includes preparing for, and executing, more effective responses:
First, we have a comprehensive staff training program to rehearse a predefined set of responses to situations that might arise in the data center, such as a utility power failure, a loss of city water, or a control system failure. We create a set of emergency response procedures for these situations, as well as other situations, and we conduct mock drills to rehearse these response procedures.
Second, we identify a crisis management team. This team is responsible for the internal and external communications around the data center event – for keeping in touch with our clients to keep them apprised of the situation. It’s important that the communications team is a separate team from the team managing a response to the event; you want the response team to be totally focused on managing the event.
Third, before we proceed to repair the problem, we make sure we understand the potential risks that might arise as a result of undertaking the repair. You need to fully understand these risks, so that you don’t damage any of the equipment, or do anything that might impact safety.
Fourth, we perform a root cause analysis, and take action to make sure the situation doesn’t happen again. Maybe someone wasn’t following procedures properly and needs additional training. Maybe there is a design issue. You need to understand these root causes, correct them, and share this information across the entire portfolio.
This interview was filmed as part of a seven-video series entitled “7 Winning Strategies for Building a Successful Data Center Business,” produced by Schneider Electric in conjunction with Datacenter Dynamics.
Jim Palmieri, Vice President of Technical Operations