Self-tuning statistical resource allocation for multipoint network events

DETAILED DESCRIPTION The present invention solves the above problem by providing a method for dynamically allocating MCU resources during a multipoint network event.
This occurs by determining the number of MCU resources to allocate at the start of the multipoint network event and then at each of a plurality of modeling intervals during the multipoint event adjusting the number of allocated MCU resources based upon the number of actual inbound users.
This method more efficiently utilizes MCU ports for a multipoint network event by allocating less than or equal to the maximum number of ports to start and then continually adjusting the number allocated based upon actual inbound users to the event.
A further method is presented for self-tuning the allocation of MCU resources for multipoint events in advance of use thereby providing a look ahead allocation of resources based on what is likely to be needed in the future for a multipoint network event.
This is accomplished by counting the number of multipoint events that have been accumulated within at least one tuning interval, accumulating a number of MCU resources actually utilized during each multipoint event in each tuning interval, and then determining a probability value for future use of MCU resources for an upcoming multipoint event based on the steps of counting and accumulating.
In one embodiment, the number of resources are accumulated during modeling intervals of each multipoint event.
The present invention provides allocation of MCU resources based on self-tuning statistics, provides configurable allocations, and provides a degree of confidence in the start parameters.