Method and apparatus for high availability distributed processing across independent networked computer fault groups

An apparatus for high availability processing comprising:

• a first common fault processor group that comprises a plurality of processors, including:
• a boot processor that includes a non volatile storage element that stores a boot program;
• and
• at least one satellite processor that boots from the boot processor;
• a second common fault processor group that comprises a plurality of processors, including:
• a boot processor that includes a non volatile storage element that stores a boot program;
• and
• at least one satellite processor that boots from the boot processor;
• wherein the first common fault processor group and the second common fault processor group are coupled together by a network;
• wherein the boot processor in the first common fault processor group is arranged to serve as a backup boot processor to the at least one satellite processor in the second common fault processor group and the boot processor in the second common fault processor group is arranged to serve as a backup boot processor to the at least one satellite processor in the first common fault processor group;
• a third common fault processor group that comprises a plurality of processors, including:
• a boot processor that includes a non volatile storage element that stores a boot program;
• and
• at least one satellite processor that boots from the boot processor;
• a fourth common fault processor group that comprises a plurality of processors, including:
• a boot processor that includes a non volatile storage element that stores a boot program;
• and
• at least one satellite processor that boots from the boot processor;
• wherein the third common fault processor group and the fourth common fault processor group are coupled together by a network;
• wherein the boot processor in the third common fault processor group is arranged to serve as a backup boot processor to the at least one satellite processor in the fourth common fault processor group and the boot processor in the fourth common fault processor group is arranged to serve as a backup boot processor to the at least one satellite processor in the third common fault processor group;
• an application that has a plurality of instances that execute on a subset of the plurality of processors in one of the first common fault processor group and the second common fault processor group and execute on a subset of the plurality of processors in one of the third common fault processor group and the fourth common fault processor group;
• and
• wherein the first common fault processor group, the second common fault processor group, the third common fault processor group and the fourth common fault processor group are coupled together by a network.

A method for providing high availability processing for an application, the method comprising the steps of:

• providing a plurality of common fault processor groups coupled together by a network, wherein each common fault processor group of the plurality of common fault processor groups comprises a plurality of processors and the plurality of processors includes:
• a host processor that includes a non volatile storage element that stores a boot program;
• and
• at least one satellite processor that boots from the host processor;
• executing a plurality of instances of the application on a subset of processors of the plurality of processors, wherein the plurality of instances are distributed across at least two common fault processor groups and the plurality of instances coordinate with each other to provide the application;
• and
• each processor of the subset of processors monitoring execution of the subset of processors to detect any failures;
• and
• redistributing execution of a plurality of instances of the application in response to detecting a failure.