Quintuple-Parity RAID Decoding for Degraded Disk Fault Tolerance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current RAID architectures face inadequate fault tolerance and high computational complexity in error correction, particularly with increasing disk sizes and requirements for error tolerance in degraded modes, leading to significant data loss due to undetected disk errors.
Innovation Solution
A systematic, linear forward error correcting code with a [N, N-5, 5] type over Galois field arithmetic is implemented, using a generator matrix that adds a row to the Information Dispersal Matrix, allowing for efficient decoding with quasi-logarithmic complexity and higher fault tolerance, including correction of up to 2 errors at unknown locations and 4 known erasures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If standard RAID levels (RAID 5, RAID 6) are used for fault tolerance, then storage efficiency is maintained, but fault tolerance becomes inadequate with increasing disk sizes and degraded mode operations
Solution Approach 1:
The patent changes the fundamental parameter of error correction capability by implementing a [N, N-5, 5] code instead of traditional RAID 5/6 codes. This increases the minimum distance from 3 to 5, enabling the system to tolerate up to 2 concurrent disk failures in degraded mode and correct undetected disk errors (UDEs), thereby adapting to modern storage requirements with larger disk sizes and extended operational periods.
Solution Approach 2:
The patent segments the error correction capability into multiple layers: (1) protection against known erasures through parity strips, (2) protection against unknown errors through increased minimum distance, and (3) degradation handling through maintained error correction in reduced configurations. This segmentation allows the system to address different failure modes independently and effectively.
2Reliability
If stronger error correcting codes are implemented to increase fault tolerance, then error correction capability improves, but computational complexity increases
Solution Approach 1:
The patent replaces complex general-purpose error correction algorithms with a specialized algebraic geometry decoding algorithm tailored for the [N, N-5, 5] code structure. This substitution leverages mathematical properties (polynomial interpolation, root finding) to achieve efficient decoding with complexity independent of the number of disks, making strong error correction practical for large-scale RAID arrays.
Solution Approach 2:
The patent changes the code parameters to [N, N-5, 5] specifically to enable efficient algebraic decoding. The minimum distance of 5 and the systematic structure allow the use of polynomial-based decoding algorithms that have computational complexity independent of N, unlike traditional approaches where complexity scales with array size.
3Quantity of substance
If RAID arrays are expanded to increase storage capacity, then storage efficiency improves, but fault tolerance deteriorates due to increased probability of multiple disk failures
Solution Approach 1:
The patent changes the error correction parameter (minimum distance) from 3 to 5, which directly addresses the reliability issue in expanded arrays. This allows the system to tolerate up to 2 concurrent disk failures, making it suitable for large arrays where the probability of multiple simultaneous failures becomes significant, while maintaining high storage efficiency.
4Productivity
If degraded mode operation is allowed to maintain availability, then system availability improves, but fault tolerance deteriorates due to reduced error correction capability
Solution Approach 1:
The patent performs preliminary action by pre-configuring the [N, N-5, 5] code structure that maintains error correction capability even when disks are lost. The algebraic geometry decoding algorithm is designed to handle degraded configurations natively, so when disks fail and the array enters degraded mode, the system continues to correct errors effectively without requiring reconstruction first.
Data Source
AI summary
An error correction and fault, tolerance method and system for an array of disks is presented. The array comprises k+5 disks, where k disks store user data and 5 disks store computed parity. The present invention further comprises a method and a system for reconstituting the original content of each of the k+5 disks, when up to disks have been lost, wherein the number of disks at unknown locations is E and the number of disks wherein the location of the disks is known is Z. All combinations of faulty disks wherein Z+2×E≤4 are reconstituted. Some combinations of faulty disks wherein Z+2×E≤5 are either reconstituted, or errors are limited to a small list.