Storage System Relay Unit Partially-Operational State Transition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage systems with redundant control paths, the reliability of access to storage devices is compromised when a control device in the secondary access path fails, leading to performance degradation as access is concentrated on the primary path.
Innovation Solution
A storage system with multiple control units, each equipped with a relay unit and an operational-state controller that can transition a failing control unit to a partially-operational state, maintaining access through a usable relay unit from another control unit, thereby preserving redundancy and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a control device in the secondary access path fails, then the reliability of access to storage devices is compromised, but access concentration on the primary path causes performance degradation
Solution Approach 1:
The system dynamically transitions the failed control device from a fully operational state to a partially operational state, where it continues to function as a relay unit for other control devices while being excluded from the access path of devices that previously used it. This dynamic state change allows the system to maintain redundancy for reliability while preventing performance degradation by redirecting access loads.
Solution Approach 2:
The control device is segmented into functional components: the access controller portion is excluded from operation for devices that previously used it, while the relay unit portion continues to operate for other devices. This segmentation allows the system to preserve redundancy benefits while avoiding the performance penalty of having a failed control device in the access path.
2Reliability
If redundancy is maintained through multiple control devices, then access reliability is improved, but system complexity increases
Solution Approach 1:
Each control device is designed with multi-functionality, serving both as an access controller for its own devices and as a relay unit for other control devices. This universal design enables redundancy without requiring separate dedicated relay components, thereby reducing overall system complexity while maintaining access path redundancy.
3Reliability
If the second access path is used for redundancy, then reliability is improved, but when the control device in the second path fails, the access path becomes unusable and redundancy is lost
Solution Approach 1:
The system dynamically adapts its topology when a control device fails. The failed device transitions to a partially operational state where it remains in the network as a relay unit for other devices, dynamically adjusting the access paths to maintain redundancy and fault tolerance despite the failure.
Solution Approach 2:
The failed control device, which would normally be completely unusable and represent a loss of redundancy, is converted into a beneficial relay unit that continues to serve other devices. This transforms the harmful effect of failure into a benefit, maintaining system redundancy and fault tolerance by repurposing the failed device's relay functionality.
Data Source
AI summary
A storage system includes a storage and control units. In each of the control units: a relay unit relays access to the storage; an access controller accesses the storage through one of the relay unit in the control unit and the relay unit in another of the control units; and an operational-state controller determines whether or not the relay unit in the above-mentioned another control unit is usable when the operational-state controller detects occurrence of an abnormality in the above-mentioned another control unit, and makes the above-mentioned another control unit transition to a partially-operational state when the relay unit in the above-mentioned another control unit is usable. In the partially-operational state, at least the access controller in the above-mentioned another control unit is not in operation and the access controller in the control unit can access the storage through the relay unit in the above-mentioned another control unit.


