|
THE INCREASING NEED FOR CLIENT/SERVER CONTINGENCY PLANNING There are a great many challenges facing the contingency planning industry in 1996. The disasters of the recent past have forced these issues to the surface and it is up to us as professionals in our field to work together to resolve these challenges. One area that is coming to the forefront of corporate contingency planning is the need to safeguard a company's LAN/WAN and PC based technologies. The proliferation of Client/Server based systems has spread the risk (to a certain extent) away from mainframe and/ or mid-range computers but has put additional pressure on servers and desktop PCs. In the past, the focus has been on recovering the BIG computer. Typical arrangements often include the use of a third party "hot-site." Increasingly, these hot-sites are also providing basic PC workstation accessibility but do not address the need for a company to recover its critical LAN environments. These "local" environments are becoming increasingly complex and important. One large southern California company has increased the number of their file servers over the past year from 5 to 8. This is projected to grow to 18 in 1996. In addition, more end-users are becoming adept at developing their own custom databases, programs, spreadsheets, etc. and are increasingly dependent on this technology for their day-to-day functions. KEY ISSUES IN THIS AREA INCLUDE: Typically, network managers rely on tape backup and RAID to safeguard their systems. Furthermore, organizations have not relied on hot-site providers since these systems are lower in cost (relative to the BIG system), are more readily available, and have not been as mission critical. In essence the strategy has been to replace destroyed systems and restore from tape. However, as file servers become increasingly customized to fit end-user requirements and the demands of end-users for instant access grows, the strategy of replacing and configuring key file servers is becoming less attractive. Solutions for the future will likely include: developing increasingly elaborate data (and voice) transmission recovery strategies, hot-site providers increasing the availability of LAN/WAN recovery programs, electronic vaulting, remote site mirroring, distributed backup servers, vendor contingency agreements, and conditioning end-users to properly document, backup, and store (off-site) critical databases and files. For example: Data transmission recovery strategies. All organizations are changing from local to nationwide to global enterprises. Therefore the need for high speed and reliable data (and voice) systems is of primary concern. In fact, many (if not all) information recovery strategies rely on these capabilities. Hot-site recovery, electronic vaulting, mirroring, and use of distributed servers all depend on the availability of data communications. The challenge for the future will likely be shaped by developing technologies, increasing efficiencies in existing systems, and the continuing fight between service providers brought on by deregulation. Hot-site providers need to look beyond mainframe and mid-range computer recovery. The geometric rate of technology advancement has changed the corporate information systems environment completely. Third party hot-site recovery providers need to look to the future and provide organizations with more than basic PC recovery services. These providers must work closely with their clients to develop comprehensive LAN/WAN recovery strategies. Electronic vaulting of critical data to a third party provider or to an organization's own equipment at an alternate site is also an option. This involves storing data en mass. When required, the data is then re-transmitted back to restore an organization's servers—once they have been repaired or replaced. Remote site mirroring, i.e. duplicating systems at an alternate location and continually transmitting data in a nearly "real time" mode. This would provide an organization with instant recovery from a remote site without the need for obtaining replacement equipment (during a disaster) and re-configuring systems. Distributed servers entails physically separate locations for an organization's key servers. This would spread the risk of one disaster destroying the entire system. Commonly, all an organization's servers are placed at the same location and may also co-exist with the mainframe and/or mid-range computer. While data centers are normally well protected against such events as fire, other events that effect the building structure and/or local access may severely impact the data center. Potential events include earthquakes, hostage/takeovers, hazardous materials spills, and civil disturbances. Vendor contingency agreements would require key vendors to provide equipment in the event of a disaster. This equipment would be used to replace an organization's servers with replacement units that are configured to the organization's pre-determined requirements. Conditioning end-users to properly document, backup, and store critical databases and files is probably the most difficult strategy for an IS Department to implement. Users that develop their own programming must adequately document their efforts. In the event of a disaster or if the user leaves the organization, this documentation will allow the IS Group to restore and modify these programs. In addition, users must also be accountable for properly backing up critical data on their local hard drives. In many organizations, PC backup only becomes a priority after a key system crashes or a disaster. Educating and communicating effectively with users will be an integral function of the corporate IS Department of the future. Developing economical and practical contingency plans will require adoption of one or more of these techniques to adequately safeguard a company's key systems. In the future, contingency planning practitioners will be depended on to develop these vital recovery strategies. About the author: |