Business Continuity Planning is a natural outgrowth of Year 2000 planning activities recently completed. The same discipline used to survive major date changes can be applied to other risk sources as well.

The first pillar of BCP is disaster recovery planning. Our experience, besides recent Y2K testing and validation, harks back to mainframe and client/server backup and recovery efforts. The main effort is to have a backup site, have multi-hosting, or have a "hot" site available. Of course, the physical backup site is only the last link in a chain of planned events. Proper execution of the planning and installation can save hundreds of manhours in building a cost-effective BCP strategy.

For a bank, we built up and tested backup arrangements with other banks. Unlike just paper plans, we executed the agreement regularly on nights and weekends to prove that the system functioned, learning, for example, that systems alone are not enough but need also paper documentation, phone access, up to date call lists, and so forth. It was fortunate that so much effort went into the planning, since the bank’s primary mainframe computer failed intermittently over a 3-week period, necessitating extensive on-site testing and parts replacement by the vendor. The logistics of running at two backup sites and a crippled main site, for a 3-week period while the main computer was being diagnosed and repaired.

For two brokerage houses, we built and extended testing laboratories for client/server applications. Besides just functionality, systems, stress, and regression testing, we designed and built component failure interrupt analysis to determine the effect and switchover to alternate paths, both for servicers and for hardware platforms. Risk events, the probability of occurrences, and the mitigation and avoidance strategies were explored and tested.