Mirrored Storage Arrays with Secondary Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current RAID systems face challenges in achieving both high performance and fault tolerance while maintaining cost-effectiveness, as they often require complex parity generation and hardware or software, and may not efficiently handle failures of multiple storage devices.
Innovation Solution
Implementing a dual storage device redundancy system with mirrored sets of primary storage devices and a secondary storage device, where data is stored in both primary and secondary caches, allowing for efficient data recovery and reduced complexity by avoiding parity calculations, enabling the use of lower performance secondary storage devices for additional redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If RAID level 1 is used to provide fault tolerance through mirroring, then data reliability is improved, but write performance deteriorates due to doubled write transaction rates
Solution Approach 1:
The system segments storage devices into multiple independent RAID 1 arrays, where each array contains mirrored pairs. This segmentation allows parallel write operations to different arrays, improving overall write throughput while maintaining fault tolerance within each array. The controller can issue write commands to multiple arrays simultaneously, effectively doubling write capacity compared to a single mirrored pair.
Solution Approach 2:
The patent combines multiple RAID 1 arrays into a unified storage system with a single namespace. The controller merges the capacity of multiple mirrored pairs into one logical volume, allowing data to be distributed across arrays while presenting a single storage interface to hosts. This merging enables the system to achieve both the reliability of mirroring and the throughput of multiple devices.
2Productivity
If RAID level 0 is used to improve I/O performance through striping, then data access speed is improved, but fault tolerance deteriorates as even one failure causes data loss
Solution Approach 1:
The system employs asymmetric redundancy where odd-numbered arrays contain full mirrored copies of data while even-numbered arrays store only a portion of the data. This asymmetric distribution allows the system to tolerate failures in even arrays without losing data, while maintaining the performance benefits of striping across all arrays. The asymmetry enables fault tolerance without requiring complete mirroring of all devices.
Solution Approach 2:
Instead of requiring complete mirroring across all storage devices, the system applies partial redundancy by storing full copies only in odd arrays and partial copies in even arrays. This partial action approach provides sufficient fault tolerance to survive single failures while reducing the write overhead compared to complete mirroring across all devices, thus improving write performance.
3Reliability
If RAID level 0+1 or 1+0 is used to provide both striping and mirroring, then fault tolerance and performance are improved, but device complexity increases and minimum four storage devices are required
Solution Approach 1:
The system creates a universal storage architecture where all arrays use the same RAID 1 mirroring mechanism, but the controller intelligently distributes data across arrays with different redundancy levels. This multi-functionality allows the system to provide both striping and fault tolerance using a single, consistent array type rather than requiring separate RAID 0 and RAID 1 configurations, simplifying management and reducing complexity.
Solution Approach 2:
The patent introduces a new dimension of organization by arranging arrays in an alternating odd-even sequence rather than grouping all mirrored pairs together. This dimensional reorganization allows the system to achieve the benefits of RAID 0+1 or 1+0 (fault tolerance plus striping) while using a simpler uniform RAID 1 configuration throughout, reducing the complexity of array management and configuration.
4Reliability
If parity-based RAID levels (3, 4, 5, 6) are used to provide fault tolerance, then data reliability is improved, but write performance deteriorates due to parity calculation requirements
Solution Approach 1:
The system uses simple copying (mirroring) instead of complex parity calculations to achieve fault tolerance. By writing identical data to multiple arrays simultaneously, the system eliminates the computational overhead of parity generation and XOR operations. This copying approach provides the same fault tolerance as parity-based systems but with significantly improved write performance, especially for random write workloads.
Solution Approach 2:
The system treats individual arrays as replaceable units where failure of any single array does not compromise overall data integrity. By using simple mirroring rather than distributed parity, the system enables faster array replacement and recovery, effectively treating failed arrays as disposable components that can be quickly replaced without complex parity recalculation, thus improving system availability and write performance.
Data Source
AI summary
A plurality of arrays of storage devices, each providing dual storage device redundancy, is provided. The plurality of arrays of storage devices includes a plurality of mirrored sets of primary storage devices, each including an even number of at least two or more primary storage devices. Each of the mirrored sets of primary storage devices stores a first and a second copy of data. The plurality of arrays of storage devices also includes a secondary storage device, which is a single physical storage device that stores a third copy of the data stored on each of the plurality of mirrored sets of primary storage devices. The secondary storage device has at least the capacity to store the data stored on the plurality of mirrored sets of primary storage devices. Dual storage device redundancy preserves data if data cannot be read from one or two physical storage devices in any array.