Critical engineering expert Kevin Rosen of Stream Data Centers discusses the importance of disaster planning in maintaining uptime and continuing mission-critical operations during and after a crisis.
Kevin Rosen, Director of Critical Engineering for Stream Data Centers
The list of possible disasters seems endless: flood, fire, cyberattack, hurricane, terrorism, blackout and more. Whether natural or man-made, events that negatively impact your business are bound to occur. And your organization faces the dual challenge of continuing operations in the midst of risks to life and property in the disaster area.
Disaster planning can minimize vulnerability by providing structured instructions on what to do in the face of threats to your data center. Whether you are preparing your first disaster plan or reviewing your current one, here are some considerations to help you develop a comprehensive plan.
Assess Your Risk
A risk assessment starts with looking objectively at what disasters are most likely to affect your region. Coastal cities can expect floods and hurricanes. Centrally located cities may have severe storms and tornadoes. West Coast areas face earthquakes and wildfires. Metro areas are subject to power blackouts or catastrophic terrorist attacks. Certainly, any kind of disaster can happen almost anywhere, but look first at your greatest risks and how likely they are to happen. Rate them according to the potential severity of such a disaster on your operations in terms of disruption, damage and economic impact.
Explore Resources
Reach out to other facilities similar to yours to find out what worked for them during crises. Take advantage of online resources to explore what happened to businesses that experienced disaster. Bring in knowledgeable data center consultants. Be willing to learn from others so you can avoid their mistakes.
Document Each Element of the Recovery Plan
Consider each risk on your worst-case scenario list to determine the path to recovery in these categories:
- Leadership — The assigned team takes responsibility for mitigating risks, preventing predictable disruptions and responding immediately to problems. This group executes the crisis communications plan in the first stages of the disaster — a crucial part of aligning the recovery effort.
- Prevention — Outline strategies to reduce the risk of damage or limit negative impact even as you locate and design your operational facility. Prevention requires experience and expertise in engineering, security, health and safety and takes all aspects of data center operation into consideration: construction, customer obligations, geographical hazards, power supply, regulatory compliance and site access.
- Evaluation — Specify how data center personnel can determine when a problem is serious enough to trigger the disaster plan, including assessment of damage to the building, power interruption, employee safety, fire potential and security.
Different data center ownership and management structures may require that a disaster recovery plan be a cooperative effort between data center owners, tenants and management.
- Response — Establish a step-by-step process for what happens with the disaster response plan is activated, especially in terms of protecting personnel and reducing damage. Develop a clear evacuation plan and consider communication necessities including activation of the emergency command bridge, personnel roles, emergency contact information, notification of customers and backup facilities, power backup, assignments to unaffected personnel, and the contingency plan for unexpected events.
- Restoration — Document the recovery process to return to normal service, including facility and road repair, equipment testing, working condition safety, staff needs, and evaluation of the successes and failures of the disaster response plan.
Practice to Make Perfect
Even if you are confident that the plan covers every likely disaster, gaps can show up between the written plan and its execution. Practice the entire plan on a regular basis, just like a fire drill. Involve the entire team, including those outside the data center. Make sure every person involved has a hard copy of the plan and contact information for the entire team. The more you practice, the greater the likelihood that the disaster response will be as seamless as daily operations, with a smooth transition to recovery mode.
Collaborate with Stakeholders
Different data center ownership and management structures may require that a disaster recovery plan be a cooperative effort between data center owners, tenants and management. Approach planning as a collaboration to develop a strategy that ensures business continuity for all companies with an interest in the facility. One size does not fit all.
Know When to Return to Business as Usual
Identifying when the disaster is over may not be as straightforward as you think. In fact, the effects of the initial event might be more devastating than the disaster itself. Have a plan for checking with people outside the area as well as local officials to determine the right time to transition away from the plan. Be patient and cautious to keep your recovery on track.
The bottom line is this: Disaster planning is not optional, unless you’re willing to put your assets and operations at risk. The time and resources spent preparing for the worst can help your mission-critical business functions emerge unscathed.
Kevin Rosen is Director of Critical Engineering for Stream Data Centers.
Stream Data Centers provides premium data center solutions, facilitated by exceptional people and outstanding service, to Fortune 500 companies.
Data Center Frontier is edited by Rich Miller, the data center industry’s most experienced journalist. For more than 20 years, Rich has profiled the key role played by data centers in the Internet revolution.