RAID Data Rebuilding with Dynamic Disk Boundary Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RAID-based storage systems face limitations in managing disk groups due to fixed fencing mechanisms, leading to unbalanced load distribution, inefficient expansion, and reduced rebuilding efficiency, especially when dealing with dynamic pool objects and proactive copy failures.
Innovation Solution
A dynamic fencing mechanism using a 'dynamic disk boundary window' and RAID extent span is introduced, allowing flexible management of disks and expanding the rebuilding write domain by optimizing disk correlations and weights, ensuring efficient data rebuilding across any number of disks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If fixed fencing mechanisms are used to manage disk groups in RAID systems, then system stability and simplicity are maintained, but rebuilding efficiency and load distribution balance deteriorate
Solution Approach 1:
The patent applies dynamics by replacing fixed fencing mechanisms with a dynamic disk boundary window approach. The boundary window dynamically adjusts based on disk correlations and RAID extent spans, allowing the system to adapt to changing data distribution patterns and optimize rebuilding operations in real-time, thereby improving rebuilding efficiency without requiring complex manual configuration.
Solution Approach 2:
The patent changes key parameters including disk boundary definitions, correlation thresholds, and window sizes to optimize rebuilding performance. By adjusting these parameters dynamically based on disk correlations and RAID configurations, the system achieves better load distribution and rebuilding efficiency while maintaining manageable complexity through automated parameter optimization.
2Productivity
If conventional RAID disk group management is used, then ease of operation is maintained, but load distribution balance and expansion efficiency deteriorate
Solution Approach 1:
The patent implements self-service by enabling the system to automatically manage disk boundaries and optimize expansion based on disk correlations and RAID extent spans. The dynamic boundary window mechanism autonomously adjusts disk group configurations during expansion operations, eliminating the need for manual intervention while achieving balanced load distribution and efficient expansion.
Solution Approach 2:
The patent achieves universality by creating a dynamic disk boundary window mechanism that handles multiple scenarios including disk expansion, failure recovery, and load balancing. This single mechanism serves multiple functions that were previously handled by separate conventional RAID management processes, improving expansion efficiency while maintaining ease of operation through unified automated control.
3Reliability
If fixed disk boundaries are used in RAID systems, then system simplicity is maintained, but rebuilding write domain size and data safety deteriorate
Solution Approach 1:
The patent applies dynamics by replacing fixed disk boundaries with a dynamic boundary window that adapts to disk correlations and RAID configurations. This dynamic approach expands the rebuilding write domain by optimizing boundary positions based on real-time disk status and data distribution, thereby reducing data loss risk while managing complexity through automated boundary adjustment.
Solution Approach 2:
The patent introduces another dimension to boundary management by incorporating disk correlations and RAID extent spans into the boundary calculation. Instead of using only positional boundaries, the system considers multi-dimensional factors including data distribution patterns, disk relationships, and RAID configurations, thereby expanding the rebuilding write domain and improving data safety through more comprehensive boundary optimization.
Data Source
AI summary
A method includes: determining, in response to a failure of a disk where a RAID extent (RE) in a storage system is located, a first disk set based on a predetermined range and the failed disk, wherein the first disk set comprises a plurality of disks within the predetermined range that are associated with the RE; determining a second disk set for rebuilding data located on the failed disk based on other disks in the first disk set than the plurality of disks that are associated with the RE; determining scores for disks in the second disk set based on disk correlations, RE spans, and weights of the plurality of disks in the second disk set; determining a target disk among the plurality of disks in the second disk set based on the plurality of scores; and rebuilding the data located on the failed disk on the target disk.