Cross-Controller Data Mirroring for Storage Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distributed storage systems face reliability issues when both a storage controller storing target data and its mirror fail, leading to interrupted host services and reduced system availability.
Innovation Solution
Implement a method where data is written into both the target controller and a mirror controller in a different control device, ensuring data redundancy across devices, and utilize a primary controller to manage mirroring relationships and establish new mirror controllers when failures occur.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is written only to the target controller and its local mirror, then writing speed is fast, but system reliability is reduced when control device fails
Solution Approach 1:
The patent segments the data writing process into two independent paths: local writing to target controller memory and remote writing to mirror controller memory in a different control device. This segmentation allows the system to maintain reliability by distributing data across control devices while preserving fast local write performance.
Solution Approach 2:
The patent introduces a spatial dimension by writing data not only to the local control device but also to a remote control device. This dimensional expansion from single-device to multi-device storage ensures that if one control device fails, data remains accessible through the remote device.
2Reliability
If data is distributed across multiple control devices, then system availability is improved, but data consistency management becomes more complex
Solution Approach 1:
The patent implements a mirroring mechanism where the target controller creates a copy of data in the mirror controller's memory. This copying approach ensures data redundancy across control devices while maintaining a simple one-to-one mirroring relationship that is easier to manage than complex multi-device distribution.
Solution Approach 2:
The patent employs a failure recovery mechanism where, upon detecting control device failure, the system automatically switches the host to access data through the mirror controller. This recovering process ensures continuous service availability without requiring complex manual intervention or data migration.
3Reliability
If write operations must complete before host receives confirmation, then data safety is ensured, but host service response time increases
Solution Approach 1:
The patent performs preliminary action by writing data to both control devices before the host receives confirmation. The target controller initiates both local and remote write operations simultaneously, ensuring data is safely stored in advance while the host waits for confirmation, thereby preventing service interruption during failure scenarios.
Data Source
AI summary
A method for improving reliability of a storage system and a related apparatus, where the storage system includes a first control device and a second control device. The method includes receiving, by a target controller, a write request, where the write request includes to-be-written data, and the target controller belongs to the first control device; writing, by the target controller, the to-be-written data into a memory of the target controller; and writing, by the target controller, the to-be-written data into a memory of a mirror controller of the target controller, where at least one mirror controller belongs to the second control device.