Parity Redistribution in Distributed RAID Arrays

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing RAID systems face challenges in efficiently redistributing parity blocks across disks when a new disk is added to a distributed parity disk array, particularly in maintaining even distribution and avoiding the assignment of multiple parity blocks from the same stripe, which complicates data recovery and storage management.

Innovation Solution

The method involves calculating a new repeat interval and creating a new parity pattern by transferring a specific number of parity blocks from each original disk to the new disk, ensuring even distribution and minimizing disruption, while avoiding the selection of multiple parity blocks from the same stripe, using a processor to execute software that manages the redistribution within the distributed parity disk array.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If parity blocks are distributed evenly across all disks in the array, then read and write performance is improved, but device complexity increases when a new disk is added to the array

Engineering Contradiction:
Improveread and write performanceVSAvoidcomplexity of redistributing parity blocks
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the parity redistribution process into distinct phases: identifying the new disk, selecting specific parity blocks to transfer, and redistributing them systematically. This segmentation transforms a complex monolithic operation into manageable discrete steps, reducing the perceived and actual complexity of the redistribution process while maintaining even parity distribution across all disks including the new one

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by first identifying and selecting which parity blocks to transfer from existing disks before actually executing the transfer to the new disk. This preliminary selection and planning phase allows the system to prepare the redistribution operation in advance, minimizing disruption to ongoing I/O operations and simplifying the actual execution phase

Inventive Principle:
Principle #10Preliminary action

2Quantity of substance

If multiple parity blocks from the same stripe are assigned to the new disk, then storage capacity utilization is improved, but data recovery reliability deteriorates

Engineering Contradiction:
Improvestorage capacity utilizationVSAvoiddata recovery reliability
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The patent implements feedback mechanisms by continuously monitoring the stripe structure and parity block distribution during the redistribution process. The system uses this feedback to make intelligent decisions about which parity blocks to transfer, ensuring that no two parity blocks from the same stripe are assigned to the new disk. This feedback-driven approach maintains data recovery reliability while still achieving effective utilization of the new disk's storage capacity

Inventive Principle:
Principle #23Feedback

3Device complexity

If a dedicated parity disk is used in RAID levels 3 or 4, then device complexity is reduced, but productivity deteriorates due to the parity disk becoming a performance bottleneck

Engineering Contradiction:
Improveparity disk configuration simplicityVSAvoiddata throughput
Core Design Contradiction:
Device complexityVSProductivity

Solution Approach 1:

The patent applies segmentation by dividing the parity storage function across multiple disks rather than concentrating it on a single dedicated parity disk. This segmentation of the parity function eliminates the performance bottleneck while maintaining manageable complexity through systematic distribution patterns that are automatically maintained during redistribution operations

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic parity distribution that adapts when disks are added or removed from the array. The system dynamically recalculates and redistributes parity blocks to maintain even distribution across the current set of disks, allowing the parity configuration to evolve with the array while preserving both performance and operational simplicity

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS8910030B2Redistributing parity in a storage system
Publication Date: 2014.12.09 NETAPP INC
  • US8910030B2 patent drawing
  • US8910030B2 patent drawing
  • US8910030B2 patent drawing

AI summary

A parity pattern defines a repeated distribution of parity blocks within a distributed parity disk array (“DPDA”). The parity pattern identifies on which disks the parity block or blocks for a stripe are located. When a new disk is added to the DPDA, the parity pattern is modified so that the distribution of parity blocks within the parity pattern is even. Parity blocks within the DPDA are then redistributed to conform with the modified parity pattern.