Live Migration Metadata Queries With Recursive Granularity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Live migration processes in data storage systems face significant computing resource drain, bandwidth consumption, and latency issues due to the creation and transfer of large metadata structures like BMPs and BDPs during data migration.

Innovation Solution

Implement a recursive large-to-small-granularity query pattern for metadata communication, using a self-describing return structure with accumulated-results data structures to reduce administrative commands and optimize bandwidth and latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional metadata communication methods are used for live migration, then complete data migration accuracy is achieved, but computing resource drain and bandwidth consumption increase significantly

Engineering Contradiction:
Improvedata migration accuracyVSAvoidcomputing resource drain
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent extracts only the essential metadata information needed for live migration using a compact bitmap structure, eliminating unnecessary data transfer. The bitmap efficiently represents dirty blocks and their offsets, allowing the system to transfer only critical migration information rather than complete metadata datasets, thus reducing computing resource drain while maintaining migration accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter representation from traditional detailed metadata formats to a compressed bitmap format with specific bit patterns. This parameter transformation reduces the amount of data that needs to be communicated and processed, directly addressing the computing resource drain issue while preserving the necessary information for accurate data migration.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If traditional metadata communication methods are used for live migration, then complete data migration accuracy is achieved, but bandwidth consumption increases significantly

Engineering Contradiction:
Improvedata migration accuracyVSAvoidbandwidth consumption
Core Design Contradiction:
ReliabilityVSLoss of energy

Solution Approach 1:

The patent extracts only the essential metadata information needed for live migration using a compact bitmap structure, eliminating unnecessary data transfer. The bitmap efficiently represents dirty blocks and their offsets, allowing the system to transfer only critical migration information rather than complete metadata datasets, thus reducing bandwidth consumption while maintaining migration accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent merges multiple metadata components into a single integrated bitmap structure that simultaneously encodes block cleanliness status, offset information, and migration requirements. This consolidation reduces the total data volume that needs to be transmitted over the network, directly addressing the bandwidth consumption issue while preserving all necessary information for accurate data migration.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If traditional metadata communication methods are used for live migration, then complete data migration accuracy is achieved, but latency increases significantly

Engineering Contradiction:
Improvedata migration accuracyVSAvoidlatency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts only the essential metadata information needed for live migration using a compact bitmap structure, eliminating unnecessary data transfer. By removing redundant metadata and transmitting only the critical bitmap information representing dirty blocks and offsets, the system significantly reduces the time required for metadata communication while maintaining migration accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary metadata processing and bitmap generation at the source storage device before initiating data migration. This advance preparation consolidates and optimizes the metadata structure in advance, reducing the processing time and latency during the actual migration operation while ensuring accuracy is maintained.

Inventive Principle:
Principle #10Preliminary action

4Productivity

If large metadata structures like BMPs and BDPs are transferred during data migration, then migration completeness is ensured, but computing resource burden increases

Engineering Contradiction:
Improvemigration completenessVSAvoidcomputing resource burden
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent changes the parameter representation from traditional detailed metadata formats to a compressed bitmap format with specific bit patterns. This parameter transformation reduces the amount of data that needs to be communicated and processed, directly addressing the computing resource burden issue while preserving all necessary information for complete and accurate data migration.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4354271B1Communicating information for performing a live migration
Publication Date: 2025.12.31 SAMSUNG ELECTRONICS CO LTD
  • EP4354271B1 patent drawingFigure 1
  • EP4354271B1 patent drawingFigure 2
  • EP4354271B1 patent drawingFigure 3

AI summary

Provided is a method for communicating information performed by a storage device, the method including determining that a first region (R1-R4) of a source storage includes a storage location including data to be copied from the source storage to a target storage, the first region (R1-R4) having a size that is equal to a first granularity size (G1), determining that a second region (R31-R44) within the first region (R1-R4) includes the storage location, the second region (R31-R44) having a size that is equal to a second granularity size (G2) that is less than the first granularity size (G1), determining that the second region (R31-R44) is associated with a threshold value, exiting a query pattern that includes the first granularity size (G1) and the second granularity size (G2), and sending a results data structure indicating that the second region (R31-R44) comprises the storage location.