System and method for parallel graph search utilizing parallel structured duplicate detection

A system for performing a graph search comprising:

• a computing system having at least a computing section, an I/O and a memory, the computing section programmed to,
• use state-space abstraction to form an abstract state-space graph that is used to partition a graph into multiple independent regions that do not interfere with each other for a period of time during the search;
• and
• use parallel structured duplicate detection to distribute a search work load across multiple processing units to maximize an amount of time during which the independent regions are capable of being maintained as independent, wherein searching of the multiple independent regions of the partitioned graph are undertaken in parallel.

A system for performing a search of a graph comprised of at least a plurality of nodes partitioned into multiple independent regions as part of a state-space abstraction operation which forms a plurality of abstract nodes, comprising:

• a plurality of processing units configured to use parallel structured duplicate detection to distribute a search workload across the plurality of processing units, allowing at least two of the plurality of processing units to search separate regions of the independent regions in parallel with each other;
• a plurality of personal memory areas in selective operational association with the plurality of processing units;
• and
• a central memory area in selective operational association with the plurality of personal memory areas, wherein the plurality of personal memory areas obtain additional memory locations from the central memory area when memory locations of the personal memory area reach a predetermined use limit.