RAID Disk Allocation for Capacity and Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage systems based on RAID, managing disk allocation effectively when the number of disks exceeds a predetermined threshold is challenging, requiring efficient division into groups to maintain high reliability and minimize data movement.
Innovation Solution
A method that determines the types and numbers of disks in multiple sets based on performance indicators, allocating disks to maximize effective capacity and reduce data movement by allocating disks in the original set to the same RAID Resiliency Sets (RRS) to minimize data relocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the number of disks in the RAID exceeds a predetermined threshold, then the storage capacity increases, but the system reliability decreases and data movement increases
Solution Approach 1:
The patent divides the RAID system into multiple RAID Resiliency Sets (RRS), where each RRS contains a limited number of disks (not exceeding the threshold). This segmentation allows the system to maintain reliability within each subset while achieving greater total capacity through the combination of multiple subsets. The division into groups ensures that failure in one RRS does not compromise the entire system.
2Reliability
If disks are divided into groups to maintain reliability, then system reliability is maintained, but data movement increases due to allocation complexity
Solution Approach 1:
The patent performs preliminary allocation of disks to RAID Resiliency Sets before actual data storage operations begin. By pre-establishing the disk groupings and RRS structures, the system avoids the need for complex real-time data movement and reorganization when adding or managing disks. This preliminary organization minimizes subsequent data movement requirements.
3Quantity of substance
If disks are allocated to multiple RRS sets, then effective capacity is maximized, but the complexity of disk management increases
Solution Approach 1:
The patent employs a systematic methodology that changes key parameters including the number of RRS sets, disks per RRS, and allocation strategies based on performance indicators. By adjusting these parameters methodically, the system maximizes effective capacity while keeping management complexity manageable through structured decision-making processes rather than ad-hoc approaches.
Data Source
AI summary
Techniques involve: in response to adding a second set of disks into a redundant array of independent disks (RAID) including a first set of disks and a total number of disks in the first and second sets exceeding a predetermined threshold, determining types and numbers of disks in the first and second sets, respectively; determining, based on the types and numbers of the disks determined and a performance indicator of the RAID having the added second set, respective types and numbers of disks comprised in a plurality of sets to which the disks in the RAID are to be allocated; and performing, based on the type and number of disks in the first set and the respective types and numbers of disks, an allocation of a disk in the first set to one of the plurality of sets, until all the disks in the first set have been allocated.