Abstract: A method and system are disclosed for maintaining a central queue for sequentially processing data within a distributed processing system. The method and system transports data to a first central queue from one or more remote processing locations for subsequent processing. Each processing location maintains in memory data transported to the first central queue as well as information relating to the data transported. Upon the processing location associated with the first central queue experiencing a failure, a second central queue is created from the data maintained in the memory of each remote processing location. The information relating to the data transported to the first central queue is utilized so that the contents of the second central queue substantially match the contents of the first central queue prior to the failure. In this way, a redundant or backup central queue can be formed without the expense of maintaining dual central queues during normal operation of the distributed processing system.

Abstract: A skills based telephone call distribution system is disclosed. The system determines the informational needs of an incoming caller and defines a set of skills an agent should possess to serve the caller. A search is performed for an available agent possessing the skill set. Upon the search failing to locate an available agent possessing the skill set, the amount of time the caller may expect to wait therefor is estimated. The estimated time is compared to a maximum waiting time. If the estimated time is greater than the maximum waiting time, then the skill set is redefined to include a greater number of agents who may serve the caller. A new search is performed for an available agent possessing the redefined skill set.