Creating a Disaster Recovery Plan

To determine what provisions to make for partial or complete loss of data, you need to determine the total cost of rebuilding or replacing the data your organization uses. Consider the following:

• What are the costs of reconstructing your organization's financial, personnel, and other business data?
• What does your business insurance cover with regard to replacing lost data?
• How long would it take to reconstruct your business data? How would this translate into lost future business?
• What is the cost per hour of server downtime?

There are several areas that need to be addressed in developing a comprehensive disaster recovery plan. Your plan for data protection needs to answer the following questions:

• What data do you need to back up and how often should you do backups?
• How will you protect critical computer or other hardware configuration information that is not saved during normal backups?
• What data needs to be stored on-site, and how will you physically store it?
• What data needs to be stored off-site, and how will you physically store it?
• What training is required so that server operators and administrators can respond quickly and effectively if an emergency occurs?

Test your plan for recovering and restoring your organization's critical data and keep copies of your disaster recovery plan both on-site and off-site.

Testing System Recovery Strategies

Testing is an important part of being prepared for disaster recovery. The skill and experience of the administrators and operators is a major factor in getting a failed computer or network back online with minimal cost and disruption to your business. You need IT personnel who are trained in troubleshooting problems and performing system recovery procedures.

Be sure to test recovery procedures before bringing a new server into production. Testing needs to include:

• Making sure that your Windows 2000 startup disks function correctly.
• Testing your uninterruptible power supply (UPS) on the computers running Windows 2000 Server and on hubs, routers, and other network components.
• Testing your disaster recovery plan.
• Performing full or partial restorations from your daily, weekly, and monthly backup media.

Practicing Recovery Procedures

You can use testing to try to predict failure situations and to practice recovery procedures. Be sure to do stress testing and test all functionality.

Some of the failures that you need to test include:

• Individual computer components such as hard disks and controllers, processors, and RAM.
• External components such as routers, bridges, switches, cabling, and connectors.

The stress tests that you set up need to include:

• Heavy network loads.
• Heavy disk I/O to the same disk.
• Heavy use of file, print, and applications servers.
• Heavy use by users who log on simultaneously.

Documenting Recovery Procedures

You need to develop step-by-step procedures for getting a computer or network back online after a disaster. Create an operations handbook that includes the following procedures:

• Performing backups
• Implementing off-site storage policies
• Restoring servers and the network

You should review your documentation when you make configuration changes to your computers or network. Updating the documentation is particularly important when you install new versions of the operating system or change the utilities or tools that you use to maintain your system.

© 1985-2000 Microsoft Corporation. All rights reserved.