Storage Array Logical Slices for RAID Performance Redundancy Trade-off
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RAID systems face a trade-off between data access performance and redundancy, where increasing redundancy leads to decreased performance and vice versa, limiting the ability to optimize both simultaneously.
Innovation Solution
Divide each storage device into logical slices, each protected by a different RAID method, allowing for optimal performance and protection without compromise by enabling flexible configuration of RAID levels across the array.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If RAID systems increase redundancy through data mirroring or parity methods, then data protection is improved, but data access performance deteriorates
Solution Approach 1:
The patent divides each storage device into multiple logical slices, allowing different RAID protection methods to be applied to different slices. This segmentation enables high-performance slices (with minimal or no redundancy) and highly protected slices (with mirroring or parity) to coexist within the same storage array, thus resolving the contradiction between data protection and access performance by allowing both to exist in different portions of the system simultaneously.
2Quantity of substance
If RAID systems use more storage devices to increase capacity, then storage capacity is improved, but the number of synchronized access operations increases reducing performance
Solution Approach 1:
The patent applies different RAID protection levels to different logical slices of storage devices. High-capacity slices can use parity-based RAID (RAID 5, 6, or 10) optimized for capacity, while performance-critical slices can use RAID 0 or RAID 1 optimized for speed. This local quality approach allows the system to achieve high overall capacity without forcing all data to undergo synchronized access operations across multiple devices, thus resolving the contradiction between capacity and IOPS.
Data Source
AI summary
A method and system of optimizing the performance, capacity and data redundancy in a storage system by defining the LBA range on each storage element managing that corresponding range (slice) on each device using a data protection method optimized for the performance and level of data protection required. The creation of such a storage array along with the replacement of failed elements and the expansion of the capacity through the addition of additional elements is managed is an automatic and transparent manner.