Parity Calculation Memory Offloads CPU Load in RAID Rebuilds

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional storage systems face challenges in minimizing the time required for data recovery during RAID rebuild processes, which can lead to data loss due to prolonged rebuild times and increased processor loads, especially when multiple drive failures occur.

Innovation Solution

A storage system with a control apparatus that utilizes a first memory apparatus with a calculation function to perform parity calculations, reducing the load on the storage controller and enabling early data recovery by distributing data recovery processes across multiple memory apparatuses and storage apparatuses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the control apparatus (CPU) performs parity calculations during the rebuild process, then data recovery can be achieved, but the processor load increases and response performance for normal I/O processes degrades

Engineering Contradiction:
Improvedata recovery capabilityVSAvoidresponse performance for normal I/O processes
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts the parity calculation function from the control apparatus (CPU) and assigns it to a dedicated parity calculation apparatus. This separation allows the CPU to focus on normal I/O operations while the specialized apparatus handles rebuild calculations, resolving the contradiction between data recovery capability and normal operation performance.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces a dedicated parity calculation apparatus as an intermediary component between the control apparatus and storage apparatuses. This intermediary handles the computationally intensive parity calculations during rebuild operations, preventing load conflicts with normal I/O processing while ensuring reliable data recovery.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the rebuild process is performed quickly to improve reliability, then data loss risk is reduced, but the complexity of coordinating multiple memory apparatuses and parity calculations increases

Engineering Contradiction:
Improverebuild completion timeVSAvoidsystem coordination complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the rebuild functionality into distinct components: control apparatus for coordination, dedicated parity calculation apparatus for computations, and multiple memory apparatuses for data storage. This segmentation allows each component to perform its specialized function efficiently, reducing overall system complexity while accelerating the rebuild process.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent designs the parity calculation apparatus as a multi-functional component that can serve both normal operational parity calculations and rebuild process calculations. This universality reduces the need for separate specialized hardware, thereby reducing system complexity while maintaining fast rebuild capability.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If distributed RAID is used to reduce write time for recovered data, then rebuild speed improves, but the range of adverse effects from drive failures widens across the capacity pool

Engineering Contradiction:
Improvewrite speed for recovered dataVSAvoidimpact range of drive failures
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent segments the capacity pool into distinct RAID groups with dedicated parity information, rather than distributing all data across a single large pool. This segmentation limits the impact of drive failures to specific RAID groups, preventing cascade failures across the entire system while maintaining high write speeds through parallel rebuild operations.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS9223655B2Storage system and method for controlling storage system
Publication Date: 2015.12.29 HITACHI VANTARA LTD
  • US9223655B2 patent drawing
  • US9223655B2 patent drawing
  • US9223655B2 patent drawing

AI summary

Provided is a storage system which allows reliability to be improved by recovering target data relatively early. A plurality of storage apparatuses 210 forms a parity group 230 for data recovery. An FM board 150 serving as a first memory apparatus is a large-capacity memory apparatus with a parity calculation function 1512. In carrying out a data recovery process, the FM board 150 acquires other data D0 and D1 and a parity P0 needed to recover target data D2, and carries out a predetermined parity calculation by the parity calculation function to recover the target data D2. When the recovered data D2 is written to a plurality of different FM boards 150, the data D2 becomes redundant and the data recovery process is therefore considered to be complete.