Distributed Storage Restriping via Intelligent Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing distributed data storage systems lack flexibility in altering virtual storage presentation to hosts, requiring modifications to either the host or network to accommodate changes in storage capacity or reliability, and involve intensive overhead in backing up and restoring data during restriping processes.

Innovation Solution

A distributed storage system with an intelligent storage element that encodes and decodes data across multiple logical units, allowing for self-managed and self-allocated virtual storage capacity, enabling restriping without the need for data backup and restoration, using a software system that multiplexes and demultiplexes data to efficiently reconfigure storage volumes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If data is restriped by backing up and restoring, then data can be moved to new storage configuration, but system overhead and time loss increase significantly

Engineering Contradiction:
Improvestorage reconfiguration capabilityVSAvoidrestriping time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent pre-calculates and stores striping mapping information that maps data striping positions across multiple logical units. This preliminary action enables direct data access during restriping operations without requiring full backup and restoration, significantly reducing the time loss while maintaining adaptability for storage reconfiguration

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces a mapping information structure as an intermediary between the data and the restriping process. This mapping information acts as a mediator that directs data flow during restriping operations, eliminating the need for complete backup and restoration while enabling flexible storage reconfiguration

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If data is restriped by backing up and restoring, then data can be moved to new storage configuration, but system complexity and operational overhead increase

Engineering Contradiction:
Improvestorage reconfiguration capabilityVSAvoidrestriping process complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the restriping process into independent mapping operations and data transfer operations. The striping mapping information is separated from the actual data, allowing the system to manage restriping through simple mapping updates rather than complex backup and restoration procedures, reducing operational overhead while maintaining storage reconfiguration capability

Inventive Principle:
Principle #1Segmentation

3Adaptability or versatility

If traditional restriping is used, then data can be relocated, but host or network modifications are required

Engineering Contradiction:
Improvestorage capacity expansionVSAvoidhost modification requirement
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent implements self-service restriping capability where the storage subsystem automatically manages its own data relocation using pre-stored mapping information. The system can add new logical units and reconfigure striping without requiring host software modifications or network protocol changes, making storage capacity expansion straightforward while maintaining adaptability

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS7913038B2Distributed storage system with accelerated striping
Publication Date: 2011.03.22 SEAGATE TECH LLC
  • US7913038B2 patent drawing
  • US7913038B2 patent drawing
  • US7913038B2 patent drawing

AI summary

A data storage apparatus and associated method is provided wherein a software system is resident in a memory space and is configured to encode data retrieved from a first number of logical units into a single channel in order to store the data in a second number of logical units.