Storage System Route Redirection for Control Board Failure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems have low stability due to the lack of redundancy, where a failure in the north bridge or CPU of a control board causes the entire storage subsystem to fail, affecting user hosts and disk chassis.
Innovation Solution
A method and device that reconfigures the storage system by modifying route information in a route information table to redirect connections from a failed control board to a non-failed control board, allowing the system to continue providing service even if a control board is damaged or in hibernation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a storage subsystem uses a single control board to serve user hosts and disk chassis, then the system structure is simple, but the system stability deteriorates when the control board fails
Solution Approach 1:
The patent merges multiple control boards into a single storage subsystem, allowing them to share resources and work together. When one control board fails, another can take over its functions, thereby improving system stability without significantly increasing structural complexity.
Solution Approach 2:
The control boards are designed with universal functionality to serve multiple purposes. Each control board can handle both front-end user host connections and back-end disk chassis connections, and can take over functions of failed control boards, enhancing system reliability while maintaining structural simplicity.
2Ease of operation
If the control board is directly connected to user hosts and disk chassis, then the connection is straightforward, but the system fails completely when the control board is damaged or in hibernation
Solution Approach 1:
The patent introduces an adapter card as an intermediary between user hosts/disk chassis and control boards. The adapter card maintains connection information and can redirect communications to alternative control boards when the primary control board fails, preserving service continuity while keeping connections straightforward.
Solution Approach 2:
The system pre-configures route information tables in adapter cards that store alternative paths and control board addresses. When a control board fails, the system can immediately use the pre-stored route information to redirect communications, ensuring service continuity without complex real-time decision-making.
3Device complexity
If route information is stored only in the failed control board, then the routing is simple, but communication cannot be maintained when the control board fails
Solution Approach 1:
The patent merges the route information storage function across multiple components - both the control board and the adapter card maintain route information tables. This redundancy ensures that communication paths are preserved even when a control board fails, while keeping the routing structure relatively simple through shared information.
Solution Approach 2:
The route information table is copied from the control board to the adapter card. This copying mechanism ensures that the adapter card has the necessary routing information to redirect communications to alternative control boards when the original control board fails, maintaining communication continuity without significantly increasing complexity.
Data Source
AI summary
A method for configuring a connection in a storage system is provided. A configuring device determines that the configuring device cannot communicate with a first control board, and identifies route information related to the first control board in a route information table. The route information is route information between an adapter card and the first control board. The configuring device modifies the identified route information by changing an address of the first control board in the route information to an address of a second control board.