Server hardware fault analysis and recovery
AssigneeQuanta Computer Inc.
InventorWei-Yu Chien
Life of this patent
Estimating the term…
Abstract
A method and system for automatically managing a fault event occurring in a datacenter system provided. The method includes collecting hardware fault event analysis corresponding with the hardware fault event. The hardware fault event analysis is organized into a report for a server device suffering from the hardware fault event. The method also includes processing statistical data received from the report for the server device. The method also includes performing hardware recovery based on the evaluated statistical data.
Claims17
A method for automatically managing a hardware fault event occurring in a datacenter system, the method comprising:
- collecting, from a server device, via a datacenter management system, hardware fault event analysis corresponding with the hardware fault event, wherein the hardware fault event analysis is organized into a report for the server device suffering from the hardware fault event;
- processing, via the datacenter management system, statistical data received from the report for the server device;
- identifying, via a BIOS service routine, a source of the hardware fault event;
- determining, via the BIOS service routine, whether the hardware fault event is a result of either a correctable or non-correctable error;
- and
- performing hardware recovery on the server device based on the processed statistical data to correct the hardware fault event in the server device.
A system for automatically managing a hardware fault event occurring in a datacenter system, the system comprising:
- a plurality of rack servers, each rack server including a server device;
- and
- a datacenter management system, wherein the datacenter management system is connected to the server device of each of the plurality of rack servers and is configured to:
- collect hardware fault event analysis corresponding with the hardware fault event, wherein the hardware fault event is associated with the server device of one of the plurality of rack servers, and wherein the hardware fault event analysis is organized into a report for the server device of the one of the plurality of rack servers suffering from the hardware fault event;
- process statistical data received from the report for the server device;
- identify, via a BIOS service routine, a source of the hardware fault event;
- determine, via the BIOS service routine, whether the hardware fault event is a result of either a correctable or non-correctable error;
- and
- perform hardware recovery based on the evaluated statistical data.
Dependencies are inferred from the claim text.