Fault-tolerant backup method and system of concurrent computer system administration controller

Abstract

The invention discloses a fault-tolerant backup principle and method for the system management controller of parallel computer. The parallel computer system is divided into a plurality of mainboard groups. Every mainboard group is enforced the monitoring and management task by a system management controller eSMC. Every two system management controllers eSMC form a group; wherein, one system management controller eSMC leads all the monitoring and management parts to be hung and connected on an I<2>C bus which is simultaneously connected with the other system management controller eSMC. When one system management controller eSMC is in shutdown, the other system management controller eSMC charges all the monitoring and management parts through the I<2>C bus. The invention is used for resolving the reliability problem of the system management controller in the parallel computer system and relieves the conflict between high usability and low cost of the high-performance computer system.

Description

technical field [0001] The invention mainly relates to the design field of a high-performance computer system management controller, in particular to a principle and method of a system management controller with a fault-tolerant backup function for a parallel computer. Background technique [0002] With the wide application of high-performance computer systems in important fields such as finance, e-commerce, data warehouses, aerospace, aviation, and scientific computing, users have higher and higher requirements for system availability. 7*24-hour trouble-free operation, end-user response time of no more than 3 seconds, and overall data security have become the main design goals pursued by system designers. The availability of the system comes from knowing the operating environment and status of the system. At present, most high-end computer systems use a system management controller to improve the availability of the system. The system management controller keeps abreast of...

AI Technical Summary

Problems solved by technology

[0008] The problem to be solved by the present invention is: aiming at the technical problems existing in the prior art, the present invention provides a low-cost embedded system management controller for parallel computers and its fault-tolerant backup method to solve the problem of system failure in parallel computer systems. Manage controller reliability issues while alleviating the conflict between high availability and low cost for high-performance computer systems

Images