DISASTER RECOVERY FOR THE LAN/WAN ENVIRONMENT:
Same Game. Different Rules.

By Troy Harris

Contrary to popular belief, Disaster Recovery Planning for a LAN/WAN environment is not entirely different from that of a mainframe environment. The fundamental objectives of Disaster Recovery Planning are the same regardless of the technologies and platforms that constitute the environment that is being protected. The challenges encountered when developing a recovery plan for a LAN/WAN environment correspond to those encountered when doing the same for a mainframe environment. The primary difference is the complexity created by the distributed computing environment and, therefore, the criticality of many elements that have traditionally been considered immaterial.

DETERMINING REQUIREMENTS

Because it is impossible to eliminate all impacts of all potential disasters, each organization must determine what level of recoverability is acceptable for its business compared to the associated costs. While conventional methods of determining an organization's recovery planning requirements, such as a Risk Assessment or Business Impact Analysis, are effective in the LAN/WAN environment, the process is not as simple as in the mainframe environment.

No longer is it adequate to evaluate the value of an individual system and its data in comparison to the costs required to protect and/or recover both. It is now imperative to consider the role of each individual system component, from host system to network component to individual workstation, in regard to the overall organization. As an example, while a particular security services server may not be identified as containing mission-critical information, the failure to recover it within the appropriate time frame may leave a separate, recovered system either unprotected or unusableÑeither of which could be devastating to an organization. This type of consideration is critical when determining the requirements of the organization's LAN/WAN Disaster Recovery Plan.

SYSTEMS MANAGEMENT DISCIPLINES

In the LAN/WAN environment, Disaster Recovery Planning costs (both tangible and intangible) can be controlled through the adoption and support of sound systems management disciplines:

• Backup Management
• Storage Management
• Configuration Management
• Security Management

While practically second nature in the mainframe environment, these disciplines have not been embraced by many organizations utilizing LAN/WAN technologies. Each of these individual disciplines contributes to the organization's recovery planning efforts while helping to control the costs of developing, implementing and supporting a sound Disaster Recovery Plan.

The objective of the systems management disciplines is to document, implement and support standard procedures, tools and measurements that allow the distributed computing environment to be managed effectively and efficiently. With standardization and documentation being two of the most significant challenges encountered when creating a LAN/WAN Disaster Recovery Plan, these disciplines can prove very valuable when incorporated into the planning effort. Sound systems management disciplines lay a foundation that can be utilized to create a cost-effective recovery strategy that encompasses a large variety of platforms and disaster scenarios.

THE HUMAN FACTOR

The distributed computing environment also has complicated the "Human Factor" of Disaster Recovery Planning. In the past, the critical IT functions could, in the event of a disaster, be performed by a limited number of individuals with specifically defined skills. A typical mainframe environment has only one operating system. Because of this, it is likely that a limited number of systems programmers would be capable of recovering and supporting the system following a disaster. In a distributed environment with numerous operating systems and protocols, a large number of skills may be required to reestablish and support an organization's computer systems following a loss or failure. Because these skills are likely to be represented by a large number of personnel, possibly dispersed over a large geographical area, the mobilization and coordination of the participants should be specifically addressed within the organization's Disaster Recovery Plan.

Additionally, the migration to LAN/WAN environments has facilitated the transition of traditional IT responsibilities into the user community. Many companies have made users responsible for backing up data residing on their own workstations and some have even implemented systems that are supported entirely outside the IT organization. Depending on the criticality of the data and processing involved, it may be necessary to incorporate these systems and the corresponding personnel into the LAN/WAN Recovery Plan. All organizations should implement an awareness and training program to communicate the corporate recovery strategy to users and to educate them on the roles they are expected to perform during a disaster scenario.

SOME GOOD NEWS!

While the migration to LAN/WAN environments has caused Disaster Recovery Planning to become significantly more complex than it was in the past, the distributed computing trend also has contributed to recovery planning in a number of ways that are worthy of consideration. The most obvious benefit is the decentralization of the risk of a disaster. It is no longer imminent that a loss of the organization’s primary computer, or a collection of computers, will necessitate the recovery of all critical data and functions. With the distribution of data and processing, the risks associated with losing both also have been distributed.

Similarly, the distributed computing environment has created a certain amount of system redundancy that can be leveraged effectively within the organization's Disaster Recovery Plan.

As organizations continue to increase their dependency on LAN/WAN environments, LAN/WAN Disaster Recovery Plans will only become more critical. By utilizing the fundamental concepts of Mainframe Disaster Recovery Planning, and recognizing the increased complexity of planning for the recovery of the LAN/WAN environment, organizations will successfully meet the challenge of Disaster Recovery Planning for the LAN/WAN environment.

About the author:
Troy Harris is the Recovery Coordinator for RCI, Inc., headquartered in Indianapolis, Indiana. As the world's largest provider of timeshare exchanges, RCI provides services to affiliates and members located throughout the world.