RAID Super-Slice Allocation for Faster Disk Rebuild Balance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Rebuild times in Redundant Array of Independent Drives (RAID) systems are elongated due to increasing disk storage capacity and unbalanced per-disk read/write loads, particularly during track failure reconstruction.
Innovation Solution
Creating data protection super-slices with super-parity, each formed from multiple RAID slices of width W, and distributing these super-slices across 2W+1 disks to balance read loads and minimize rebuild time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If disk storage capacity increases, then data storage capability improves, but rebuild time increases
Solution Approach 1:
The patent divides the rebuild process into parallel segments by distributing super-slice tracks across multiple disks (2W+1 disks) and enabling simultaneous read operations from multiple disks. This segmentation allows the system to rebuild failed tracks in parallel, reducing overall rebuild time despite increased storage capacity.
2Reliability
If RAID slice width increases, then data protection capability improves, but per-disk read/write load becomes unbalanced
Solution Approach 1:
The patent introduces a new dimensional organization by creating super-slices that span across 2W+1 disks, transforming the traditional single-RaidSlice structure into a multi-dimensional distribution pattern. This dimensional change allows tracks to be distributed more evenly across disks, balancing the read/write loads while maintaining strong data protection through super-parity.
3Speed
If super-slice track distribution is optimized, then rebuild speed improves, but system complexity increases
Solution Approach 1:
The patent performs preliminary track allocation and super-slice creation during the normal operation phase, organizing tracks across 2W+1 disks in advance. This preliminary action ensures that when a failure occurs, the system can immediately initiate parallel rebuild operations without needing to perform complex real-time calculations, thus achieving fast rebuild speeds while managing system complexity.
Data Source
AI summary
Data protection super-slices with super-parity p are created from sets of two RAID slices a, b of width W that each include slice parity a, b. The tracks of each super-slice are distributed across the disks of a disk cluster of width 2W+1, and the same number of indexed sequences. Each of the indexed sequences contains one super-slice with the slice tracks a, b distributed according to a fixed sequence in order from a start pointer to a stop pointer. The start pointer and the stop pointer are shifted one place for each successive indexed sequence. The fixed sequence may be characterized by slice track adjacency or be interleaved, e.g., (a b b a a a a a b b b b a b b b a a b a b b a a bap) or (a a b a b b a b b b b a a a b a a a b a a a b b b bp).