Best Practices for Data Center Disaster Recovery

Today, data center operators worry about high availability of their data centers every day, working around the clock to ensure 100% uptime. They deploy redundant facilities to minimize risk, rigorously plan and test to ensure continuous operations, and take precautions to protect their infrastructure from environmental threats. However, even the best plans and preparations can go wrong in natural disasters or emergencies.

In fact, in some extreme cases, advanced planning and preparation are still powerless, and disaster recovery becomes an important part of ensuring that enterprises can maintain functionality while rebuilding data center operations.

In any disaster situation, time is of the essence, so data center personnel need to know the appropriate actions to take in the minutes, hours, and days following an event.

[[209568]]

The safety of your personnel is the first and foremost factor, so once it is determined that all personnel are safe, the following recovery activities should begin:

• Proactively monitor all critical equipment to look for hidden problems - Look at equipment that is directly affected or exposed to the disaster event, such as generators, oil, and filters. For example, in situations where recovery efforts may take days or even weeks, it is critical to monitor fuel supplies and understand how long generators can run without backup diesel. In natural disaster emergencies, hospitals, first responders, and other public safety organizations must first obtain fuel supplies, regardless of whether the data center has a fuel supply contract or not.
• Increase site inspections during designated time periods - This will help staff identify any deficiencies such as leaks or damage to facilities. It is also important to ensure that each employee has a partner to provide support and assistance.
• Ensure staff rotation to reduce pressure on on-site staff - To provide employees with adequate rest time, companies can consider recruiting staff from other unaffected data centers or entrusting third-party vendors to carry out partial recovery work.
• Establish communications protocols to provide critical updates and field feedback - After validating all areas and systems, hold regular briefings to inform staff and management of any data center or field impacts. In a disaster recovery situation, everyone from the CIO to every staff member has a role to play, so it is imperative that all parties involved are kept up to date on what is happening.
• Monitor input power - Since the grid that provides power to the data center may be affected, the data center power will be interrupted or the quality will be reduced, so it is very important to contact the power company to understand a reasonable safe time, when the grid power supply is normal, the data center shuts down the backup generator and reconnects to the grid.
• Apply Lessons Learned – As soon as time permits, begin writing and documenting lessons learned and best practices to prepare for the next step.

While natural disasters are an inevitable risk, with the right disaster recovery plan, data center operators can minimize the impact of these catastrophic events and resume operations quickly and safely.