RAID Member Index Swapping for Low-Movement Array Growth
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RAID systems face inefficiencies in maintaining predictable distribution of protection group members and spare capacity during drive cluster scaling, leading to increased data movement and system resource engagement.
Innovation Solution
Implement a drive rotation-relocation algorithm that swaps drive indices to recreate a recurrent pattern of RAID member distribution, reducing the need for physical relocation of drives and minimizing data movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of stationary object
If traditional RAID scaling methods are used, then drive cluster capacity is increased, but data movement and system resource engagement increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-calculating and pre-organizing the distribution pattern of RAID members before actual drive addition occurs. The system determines the recurrent distribution pattern in advance and prepares the mapping between old and new drives, thereby avoiding chaotic data movement during the actual scaling operation. This preliminary planning minimizes system resource engagement during the critical data relocation phase.
Solution Approach 2:
The patent utilizes parameter changes by transforming the drive indexing parameters and redistribution rules when drives are added to the cluster. Instead of maintaining fixed indexing schemes, the system dynamically adjusts the indexing parameters and redistribution algorithms to accommodate new drives while preserving the recurrent distribution pattern. This allows seamless scaling without triggering large-scale data remapping operations.
2Stability of the object's composition
If RAID members are redistributed during scaling, then uniform distribution is maintained, but data movement increases
Solution Approach 1:
The patent applies segmentation by dividing the drive cluster into distinct segments or zones with specific distribution characteristics. When drives are added, the system segments the redistribution operation into targeted portions rather than requiring complete re-distribution across all drives. This selective segmentation maintains distribution uniformity in critical areas while minimizing unnecessary data movement in other areas.
Solution Approach 2:
The patent employs asymmetry by implementing non-uniform redistribution strategies for different types of drives or different regions of the cluster. Instead of applying symmetric redistribution rules to all drives, the system uses asymmetric mapping rules that take into account the specific roles, capacities, and positions of drives. This asymmetric approach maintains overall distribution uniformity while reducing total data movement by optimizing relocation paths.
3Stability of the object's composition
If rotation-relocation algorithm is used, then recurrent distribution pattern is maintained, but drive indexing complexity increases
Solution Approach 1:
The patent applies copying by creating and maintaining a virtual copy or model of the recurrent distribution pattern. Instead of directly manipulating complex drive indices, the system uses a simplified virtual indexing model that mirrors the actual drive structure. The rotation-relocation operations are performed on this virtual model, and the results are then mapped back to the physical drives. This copying approach maintains distribution pattern consistency while hiding the indexing complexity from the main control logic.
Data Source
AI summary
A predictable distribution of protection group members and spares in compliance with RAID requirements is maintained as a drive cluster is scaled-up and eventually split into multiple clusters. Conditions under which addition of new disks using rotation-relocation and cell borrowing would create recurrent patterns of distribution of protection group members are recognized and the associated groups of disks are index-swapped to reduce the amount of data and parity information that needs to be relocated.