Storage Control Device PCIe Abnormality Detection and Reset
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In RAID systems, accurately identifying abnormal components is challenging due to the characteristics of PCIe, leading to erroneous specification of normal components as suspicious, resulting in unnecessary shutdowns and maintenance errors.
Innovation Solution
A storage control device with paired control units, each equipped with an interface unit, detection unit, and reset control unit, which detects abnormalities and resets the interface unit of the other control unit to prevent erroneous shutdowns and maintain system operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system retracts the control unit specified as suspicious based on PCIe abnormality detection, then the reliability of the RAID system is improved by isolating the abnormal component, but normal control units may be erroneously retracted due to inaccurate abnormality specification
Solution Approach 1:
The system performs a reset operation on the interface unit of the suspicious control unit before final retraction. This preliminary action allows the system to test whether the abnormality is transient or permanent, preventing erroneous retraction of normal control units while maintaining reliability by isolating truly abnormal components
Solution Approach 2:
The system uses feedback from the reset operation results to determine whether to retract the control unit. By monitoring whether the abnormality persists after reset, the system can accurately distinguish between transient errors and permanent failures, improving both reliability and detection precision
2Ease of repair
If the system performs maintenance on erroneously specified control units, then the system undergoes unnecessary shutdowns and maintenance operations, but the underlying abnormality in the actual faulty component remains unaddressed
Solution Approach 1:
The system performs a reset operation as a preliminary step before maintenance. This allows many abnormalities to be resolved without physical maintenance, reducing unnecessary shutdowns and time loss while maintaining ease of repair for cases that do require intervention
Solution Approach 2:
The system changes the operational state parameter of the interface unit by performing a reset operation. This parameter change can resolve transient abnormalities without requiring physical maintenance, thereby reducing system downtime while maintaining the ability to perform maintenance when truly needed
Data Source
AI summary
Each CM includes an interface unit, a first detection unit, and a reset control unit. The interface unit is configured to be connected to a communication channel and control communication using the communication channel. The first detection unit is configured to detect an abnormality in an inter-CM path including the interface unit of the CM, to which the first detection unit belongs, the interface unit of the other CM, and the communication channel. The reset control unit is configured to retract the other CM and reset the interface unit of the one CM in a case where an abnormality of the other CM side is detected by the first detection unit. Accordingly, even in a case where a suspicious control unit in which an abnormality has occurred is erroneously specified, the maintenance of the suspicious control unit can be performed with the operation being continued.