Shared Reserved Extent Pool for Storage Data Rebuild

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage systems based on Redundant Arrays of Independent Disks (RAID) face inefficiencies in utilizing reserved storage areas during normal operation, as these areas are not effectively utilized for data recovery when storage devices fail, leading to suboptimal storage space allocation and management.

Innovation Solution

A method and device for managing a storage system that detects failed storage devices, determines the extent assigned for building a stripe, and selects a spare reserved extent from shared reserved extents across groups to rebuild data, allowing for efficient data recovery and allocation of storage space by using a shared reserved storage area across multiple groups.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If separate reserved areas are arranged in each storage device for data recovery, then data reliability is improved, but storage space utilization deteriorates because reserved areas cannot be utilized effectively during normal operation

Engineering Contradiction:
Improvedata reliabilityVSAvoidstorage space utilization
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The patent merges separate reserved areas from multiple storage devices into a shared reserved storage pool. Instead of each storage device having its own dedicated reserved area, the reserved extents from multiple devices are combined into a common pool that can be shared across the entire storage system, improving overall utilization while maintaining data recovery capability.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The shared reserved storage pool serves multiple functions: it acts as reserved space for data recovery when devices fail, while also being available as allocatable storage space during normal operation. This multi-functional approach allows the same storage resources to serve both reliability and capacity needs.

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

2Productivity

If reserved storage areas are arranged in each storage device to ensure quick data rebuilding, then data recovery speed is improved, but storage capacity is reduced due to duplicate reserved spaces

Engineering Contradiction:
Improvedata recovery speedVSAvoidallocatable storage space
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent combines reserved extents from multiple storage devices into a unified shared reserved storage pool. This eliminates the need for each device to maintain separate reserved spaces, thereby increasing allocatable storage space while maintaining the capability for quick data recovery through the shared pool.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If separate reserved areas are allocated for each group of storage devices, then data security between groups is improved, but storage management complexity increases

Engineering Contradiction:
Improvedata security between groupsVSAvoidstorage management complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The shared reserved storage pool serves both groups simultaneously, providing a universal resource that can be accessed by either group when needed. This single unified structure replaces the need for separate reserved areas for each group, reducing management complexity while maintaining security through controlled access mechanisms.

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

Data Source

PatentUS11829247B2Method, device, and computer program product for rebuilding data of a failed storage device in a storage system
Publication Date: 2023.11.28 EMC IP HLDG CO LLC
  • US11829247B2 patent drawing
  • US11829247B2 patent drawing
  • US11829247B2 patent drawing

AI summary

Techniques manage a storage system. Along these lines, a failed storage device is detected among storage devices included in a resource pool of the storage system, and the storage devices belong to a first group and a second group in the resource pool respectively. An extent in the failed storage device assigned for building a stripe in the storage system is determined. A spare reserved extent is selected from respective reserved extents included in respective storage devices among the storage devices based on a mapping relation between the stripe and a plurality of extents in the stripe. The respective reserved extents are shared among the first group and the second group. Data in the extent are rebuilt to the selected spare reserved extent. A reserved storage area can be shared among various storage devices in the resource pool, so as to increase allocatable space in storage devices.