Snapshot Chain Metadata Deletion for Distributed Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for deleting snapshots in distributed storage systems are inefficient due to the need for extensive data read and write operations, leading to low deletion efficiency.

Innovation Solution

A method and apparatus that utilize meta information to determine target sub-data within snapshot chains, allowing for efficient deletion by identifying and removing unreferenced data based on snapshot levels and data determination modes, enhancing system isolation through distributed object-based and block-based storage systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data read and write operations are performed on massive snapshots to determine data for deletion, then the deletion can be completed, but the efficiency of data deletion becomes low

Engineering Contradiction:
Improvedeletion accuracyVSAvoiddeletion efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent creates a copy of the snapshot chain structure (meta information copy) to perform deletion operations on the copy rather than directly on the original massive snapshot data. This allows efficient processing of deletion targets through metadata manipulation while maintaining accuracy, resolving the contradiction between deletion reliability and efficiency.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent extracts only the necessary meta information from the massive snapshot data structure, separating the essential deletion target identification data from the full snapshot contents. This extraction enables efficient processing by working with a smaller, manageable metadata representation while preserving deletion accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

2Reliability

If snapshot chains are maintained with original view at the head and oldest snapshot at the tail, then data backup and recovery are enabled, but storage space consumption increases

Engineering Contradiction:
Improvedata recovery capabilityVSAvoidstorage space
Core Design Contradiction:
ReliabilityVSVolume of stationary object

Solution Approach 1:

The patent implements a nested snapshot chain structure where snapshots are organized hierarchically with the original view at the head and older snapshots nested toward the tail. This nesting allows efficient data recovery by traversing the chain while enabling space optimization through identification and deletion of unreferenced nested snapshots.

Inventive Principle:
Principle #7Nested doll (Nesting)

Solution Approach 2:

The patent enables selective discarding of unreferenced snapshots in the chain while maintaining the recovery capability for referenced snapshots. By identifying snapshots that are no longer needed through meta information analysis, the system can discard them to free storage space while preserving the integrity of the recovery chain for active snapshots.

Inventive Principle:
Principle #34Discarding and recovering

Data Source

PatentEP3923155B1Method and apparatus for processing snapshot, device, medium and product
Publication Date: 2026.03.04 BEIJING BAIDU NETCOM SCI & TECH CO LTD
  • EP3923155B1 patent drawingFigure 1
  • EP3923155B1 patent drawingFigure 2
  • EP3923155B1 patent drawingFigure 3~4

AI summary

The present disclosure discloses a method and apparatus for processing a snapshot, a device, a medium and a product, and relates to the field of computer, and particularly to the field of cloud computing technology, distributed storage technology or cloud storage technology. An implementation of method may include: acquiring meta information of snapshots; acquiring, for each snapshot in the snapshots, a sub-data list of the each snapshot based on the meta information of the each snapshot; determining target sub-data in the acquired sub-data list according to a data determination mode corresponding to a snapshot level of the each snapshot; and deleting the target sub-data.