Methods and apparatus for handling processing errors in a multi-processing system

An apparatus, comprising:

• a plurality of sub-processing units, each operable to perform processor tasks;
• and
• a main processing unit operable to:
• (i) monitor the processor tasks and associated processor loads therefor that are allocated to be performed by the respective sub-processing units;
• (ii) detect whether a processing error has occurred in a given one of the sub-processing units;
• and
• (iii) re-allocate all of the processor tasks of the given sub-processing unit to one or more participating sub-processing units, including other sub-processing units associated with the main processing unit, based on the processor loads of the processor tasks of the given sub-processing unit and the processor loads of the participating sub-processing units;
• and
• one or more respective groups of sub-processing units, each group being associated with a respective main processing unit, wherein the participating sub-processing units include one or more of the respective groups of sub-processing units,
• wherein each of the respective groups of sub-processing units and the associated main processing unit:
• (i) are part of a set of multi-processing units, wherein the respective group of sub-processing units and the main processing unit share a common data bus with one or more others of the respective groups of sub-processing units and associated main processing units;
• (ii) are a stand alone multi-processing unit, wherein the respective group of sub-processing units and the main processing unit does not share a common data bus with any other group of sub-processing units and associated main processing unit;
• or
• (iii) are at least part of a distributed multi-processing unit, wherein at least some of the main processing unit and the sub-processing units are remotely located from one another;
• wherein at least some of the multi-processing units, the stand alone multi-processing units, and the distributed multi-processing units may be disposed at least one of:
• (i) on common or different circuit boards;
• (ii) in common or different products;
• and (iii) in common or different locations;
• the main processing unit is further operable to determine communication requirements, including at least one of communication bandwidth and communication latency, needed between the given sub-processing unit and one or more sub-processing units to share processing results with the given sub-processing unit;
• and
• the re-allocation of the tasks of the given sub-processing unit is further based on the communications requirements.