Hewlett Packard Enterprise Development LP

Silicon Graphics, Incorporated

Silicon Graphics International Corporation

Mark Larson

Michael Brown

Gary Meyer

Richard T Ingebritson

Error data is read from error registers and written into a buffer. A computing node uses a BIOS to read the error data, rearm the error register and write the data into a memory mapped buffer. A hub chip supports creation of a shared memory system of computing nodes. A management controller in the computing node extracts error data from the buffer. The error data preferably consists essentially of the error register identifiers and the contents of the error registers. A system management node receives the error data from the management controllers in the computing nodes. The system management node may be coupled to but separate from the computing nodes.

Method for error logging

System and method for error logging

Diagnostic environments designed for ccNUMA computing systems require different capabilities than conventional diagnostic programs. Since most conventional operating systems do not permit sufficient access to hardware by application programs, a diagnostic microkernal is distributed to all of the processors of all of the nodes of the computing system. Instead of dumping output to a user interface, diagnostic programs executing under control of the diagnostic microkernal store formatted data in shared memory. A shell process executing at an interface node is notified of errors and informs the user who selects the formatted data in shared memory to be accessed by the shell process for output via a user interface by little more than echoing the data. Either a dumb terminal or a desktop computer or workstation which adds further formatting capabilities may be used for display of the data.

Diagnostic system and method for a highly scalable computing system

Growing community of inventors

Eau Claire, WI, United States of America

Mark Larson