Port-Specific Reset for RAID Controller Error Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In conventional RAID storage devices, when one Controller Module (CM) fails, the error propagation to other normal CMs leads to unnecessary resetting of the secondary CM, causing it to become unusable, which disrupts the storage device's operation.
Innovation Solution
The proposed solution involves a redundant control device configuration with a switch device having separate ports connected via different channels, an error detection unit, and a reset processing unit that performs port-specific resets to isolate the faulty CM, preventing error propagation and maintaining the secondary CM's functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a reset instruction is transmitted to isolate a faulty CM, then error propagation is prevented, but the secondary CM becomes unusable due to unnecessary resetting
Solution Approach 1:
The patent segments the reset operation into port-specific resets rather than resetting the entire CM. The switch device performs reset only on the specific port connected to the faulty CM, leaving other ports and the secondary CM itself operational. This segmentation prevents unnecessary resetting of the secondary CM while still isolating the error effectively.
Solution Approach 2:
The patent applies local quality by making the reset operation localized to specific ports rather than applying it globally to the entire CM. The switch device identifies which port is connected to the faulty CM and applies the reset instruction only to that specific port, preserving the functionality of other ports and the secondary CM.
2Reliability
If connection between CMs is cut off instantly upon failure detection, then failure propagation is prevented, but the secondary CM's functionality is disrupted
Solution Approach 1:
The patent segments the connection isolation to the port level rather than the entire CM level. The switch device cuts off communication only through the specific port connected to the faulty CM, while maintaining connections through other ports. This allows the secondary CM to remain operational and maintain productivity.
Solution Approach 2:
The patent applies local quality by isolating the failure at the local port level rather than globally affecting the entire CM connection. The switch device applies the isolation instruction locally to the specific port connected to the faulty CM, preserving the secondary CM's ability to function and maintain storage device productivity.
3Reliability
If a reset instruction is transmitted to the switch device, then the faulty CM is isolated, but the entire switch device may be affected
Solution Approach 1:
The patent segments the switch device's reset capability into individual port-level resets. Instead of resetting the entire switch device, the system performs reset operations on specific ports only. This segmentation contains the error impact to the specific port level while preserving the rest of the switch device's functionality.
Solution Approach 2:
The patent applies local quality by enabling the switch device to apply reset instructions locally to specific ports rather than globally to the entire device. The switch device identifies which port is connected to the faulty CM and applies the reset instruction only to that port, maintaining the complexity management at a reasonable level.
Data Source
AI summary
A first control device includes: a switch device including a first port connected to a second control device among the plurality of control devices via a first channel and a second port connected to the second control device via a second channel and to which a processing device is connected; a detection unit that detects an error in the control devices; a first reset processing unit that performs a port reset of the first port included in the switch device; and a transmitting unit that transmits a reset instruction to the second control device; thereby propagation of an error occurred in a control device can be inhibited.