IO Array Selection Using Volume State and Array Similarity Changes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data replication and backup systems face performance and availability issues due to inflexible switching logic that does not account for real-time performance indicators or failure severity levels, leading to suboptimal IO array selection during failures.
Innovation Solution
A method for selecting an IO array based on state change information of volume dimension and similarity change information of storage array dimension, allowing dynamic and accurate determination of suitable storage arrays for IO operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system constantly switches between IO arrays to ensure no IO errors, then reliability is improved, but system performance deteriorates due to frequent switching
Solution Approach 1:
The patent implements dynamic IO array selection by continuously monitoring volume states and array states, and adjusting the IO array selection based on real-time conditions. The system dynamically determines whether to switch IO arrays based on state change information and similarity change information, rather than constantly switching. This dynamic adaptation resolves the contradiction by making the switching behavior conditional and intelligent, improving reliability only when necessary while maintaining performance during normal operation.
2Device complexity
If the system uses static IO array selection, then system complexity is reduced, but adaptability to real-time conditions deteriorates
Solution Approach 1:
The patent employs feedback mechanisms by continuously acquiring volume states and array states, comparing current states with historical states, and using the state change information to dynamically adjust IO array selection. The system monitors similarity changes between storage arrays and feeds this information back to the selection logic. This feedback-driven approach enables the system to adapt to real-time conditions without requiring overly complex switching logic, as the complexity is managed through systematic state monitoring and comparison.
Solution Approach 2:
The patent changes the parameters used for IO array selection from static configuration to dynamic state-based parameters. Instead of relying on fixed selection criteria, the system uses volume state change information, array state similarity change information, and failure severity levels as dynamic parameters. This parameter transformation allows the system to adapt to real-time conditions while keeping the switching logic structure relatively simple, as it primarily involves comparing state parameters rather than complex decision trees.
3Adaptability or versatility
If the system implements dynamic switching based on real-time conditions, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent segments the IO array selection process into distinct functional modules: volume state monitoring, array state monitoring, state change detection, similarity calculation, and selection decision-making. By dividing the complex dynamic switching system into these smaller, specialized components, the overall complexity is made more manageable. Each module has a specific function, and they can be implemented and maintained independently, reducing the burden of managing system complexity while preserving adaptability.
4Reliability
If the system moves unselected storage arrays offline, then IO error prevention is improved, but availability deteriorates as fewer arrays provide services
Solution Approach 1:
The patent applies partial action by selectively moving only the unselected IO array offline while keeping the selected IO array online and providing services. This partial offline approach prevents IO errors by isolating potentially problematic arrays from the active IO path, while maintaining service availability through the selected array. The system does not move all arrays offline or prevent all arrays from providing services, but rather applies the offline action selectively based on selection status, thus resolving the contradiction between error prevention and availability.
Data Source
AI summary
Techniques for selecting an input/output (IO) array from storage arrays involve acquiring groups of volume states of a target volume of volumes in the storage arrays, wherein each of the groups of volume states includes volume states at different time points. Such techniques further involve determining, based on the groups of volume states, pieces of state change information of the target volume in the storage arrays. Such techniques further involve determining similarities among groups of array states of the storage arrays and similarity change information among the similarities, wherein an array state of each of the storage arrays is based on a volume state set of the volumes in each of the storage arrays. Such techniques further involve selecting, based on the similarity change information and the pieces of state change information corresponding to the storage arrays, an IO array from the storage arrays which is used for IO.