Method to improve tolerance of non-homogeneous power outages

An multiprocessor system comprising:

• a network;
• a plurality of processors communicatively coupled by means of said network and configured as co-peers for purposes of reconfiguring, each of said plurality of processors having a respective memory wherein is located a computer program for causing said respective processor to reconfigure said system when a predetermined critical set of resources is necessary for operation of said multiprocessing system by
• configuring a register to represent said predetermined critical set of resources;
• after said configuring, recognizing failure of a first of said plurality of processors to receive a message periodically broadcast by a second of said plurality of processors by means of said network, said message indicating the continued operation of said second processor;
• in response to said recognition of failure, determining a set of processors, said set comprising processors of said plurality of processors that indicate their continued operation;
• and
• reconfiguring said system and continuing operation of said system and ostracizing from said multiprocessor system any processor of said plurality of processors not a member of said set only if said set of processors comprise said predetermined critical set of resources.

An article of manufacture comprising a medium for data storage wherein is located a computer program for causing a processor in a multiprocessor system to computer system to reconfigure said system when a predetermined critical set of resources is necessary for operation of said multiprocessing system by

• communicatively coupling said plurality of processors by means of a network;
• and configuring said plurality of processors as co-peers;
• configuring a register to represent said predetermined critical set of resources;
• after said coupling and configuring, recognizing failure of a first of said plurality of processors to receive a message periodically broadcast by a second of said plurality of processors by means of said network, said message indicating the continued operation of said second processor;
• in response to said recognition of failure, determining a set of processors, said set comprising processors of said plurality of processors that indicate their continued operation;
• and
• reconfiguring said system and continuing operation of said system and ostracizing from said multiprocessor system any processor of said plurality of processors not a member of said set only if said set of processors comprise said predetermined critical set of resources.