Planning for Disaster Recovery? Take these Cues

by CXOtoday Staff    Nov 11, 2008

Plan DR for entire Organization, not just IT department

Disaster recovery (DR) planning encompasses the entire organization, not simply the IT services department. Preparing for contingent facilities and labor, building communication processes for employees, establishing plans with suppliers and distributors and outlining actions to follow in order to quickly recover normal business operations are all required for successful recovery.

Plan for disaster avoidance too

“In fact, planning for disaster avoidance should be conducted in addition to DR planning. For example, implementing proper information security controls and virus/spyware protection schemes helps to avoid potential security break-in or data loss disasters, or constructing facilities that can withstand natural or man-made calamities. While DR planning is required for the whole organization, including all resources, functions, and facilities, the scope must be focused on IT systems and infrastructure: protecting the data center,” says Anand Naik, director, Systems Engineering for Symantec.

Plan for Backup utility power

Disasters damage, and in more extreme cases, completely annihilate the working environment. Even if the workplace is spared, disasters such as a fire incident can cause mayhem and take out utilities on which the workplace is dependent. Backup power systems comprising uninterruptible power supply (UPS) systems and diesel generator backup are expected core components of modern data centers, but can only supply power until the diesel fuel is spent. Therefore, plans should include methods of extending backup utility power in case of long utility outages.

Distance between sites is critical

Poorly planned architecture or ongoing construction work can result in accidents that result in cutting utilities or Internet service supplies to the workplace as well as work sites that are even geographically distant from the affected work. Therefore distance between the sites can be critical, adds Naik.

Get certified to avoid prosecution of staff

Another critical area involves meeting regulatory requirements to avoid legal prosecution of executive staff and payments of large fines. One way of going about it could be through certification of business processes to avoid disasters as well as ensure rapid recovery should disaster strike. A prime example to this would be ISO 9001-certified manufacturing processes that are increasingly becoming significant for business interaction and ISO 27001 data security certification requirements that are on the rise.

Replace traditional storage with new techniques

A good start would include ensuring traditional tape backup/restore is relegated to satisfying long-term off-site archival and data retention needs. Tape is not dead, but continues to be the lowest-cost storage medium for large data sets (terabyte or larger systems) both in hardware costs and utility costs (electricity).

“We try and encourage our customers to consider replacing older tape backup systems with VTLs, D2D, replication, and snapshot solutions for those systems suffering from data growth. These solutions are able to keep pace with data growth. Furthermore, these solutions offer significantly improved restore times for individual file or object restore requirements. However, for DR, VTL systems should be replicated to alternate locations,” says Naik.

DR with Virtual Machines

Virtual machines have recently found their way in to providing unique and cost effective DR solutions. A number of characteristics of virtual machine technology enable and simplify DR processes that include becoming a solution to the hardware version problem. Restoring from an image also exhibits similar compatibility issues. Additionally, a virtual server can be quickly provisioned without regard to drivers and server-specific configurations. Although, DR may not require all critical services to run at full production capacity initially, recovering servers and services into virtual machines offers the added benefit of reduced standby hardware, mitigating overall investment and operating costs.