Method, system and program for securing redundancy in parallel computing system

A parallel computing system having a plurality of computing node groups having respective node group information, including at least one spare computing node group for securing redundancy, comprising:

• an information management server having a plurality of managing nodes for allocating jobs to their associated computing node groups, and having respective computing node group status information;
• wherein the system is configurable using three manners of computing node group switching;
• and
• the information management server in a first manner, the first of the plurality of managing nodes in a second manner, or all managing nodes in a third manner are configured to:
• update the respective in-use computing node group status information by accessing the information management server;
• receive a job;
• queuing the job for processing on the associated computing node group;
• detect an occurrence of a failure in the associated computing node group and in response:
• identify a spare computing node group by accessing the computing node group status information in the information management server;
• obtain computing node group information of the spare computing node group;
• and
• continue processing of the queued job by switching the disabled computing node group to the spare computing node group on the basis of the computing node group information of the spare computing node group;
• detect a recovery of the disabled computing node group and in response:
• register the disabled computing node to the information management server as a new spare computing node group.

A program product for securing redundancy in a parallel computing system having a plurality of computing node groups having respective node group information, including at least one spare computing node group, the program product comprising a computer readable storage media bearing instructions for securing the redundancy in the parallel computing system by causing the parallel computing system to execute the acts of:

• associating a plurality of managing nodes with one of a plurality of computing node groups and with an information management server having respective computing node group status information for the plurality of computing node groups;
• receiving a job at a first of the plurality of managing nodes;
• queuing the job for processing on the computing node group associated with the first of the plurality of managing nodes;
• updating, by the first of the plurality of managing nodes, the in-use status information for the associated computing group node by accessing the information management server;
• detecting, by the first of the plurality of managing nodes, an occurrence of a failure in the associated computing node group;
• identifying, by the information management server in a first manner, the first of the plurality of managing nodes in a second manner, or all managing nodes in a third manner, a spare computing node group by accessing the computing node group status information in the information management server;
• obtaining, by the information management server in a first manner, the first of the plurality of managing nodes in a second manner, or all managing nodes in a third manner, computing node group information of the identified spare computing node group;
• disabling the computing group node associated with the detected failure;
• reassociating the first of the plurality of managing nodes with the identified spare computing node group;
• continuing, by the first of the plurality managing nodes, processing of the queued lob using the identified spare computing node group;
• when the failure of the disabled computing node group is recovered, switching the disabled computing node group to the identified spare computing node, and
• registering the disabled computing node to the information management server as a new identified spare computing node group.