Virtual Machine Snapshot Hierarchy Migration Across Object Stores

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Migrating virtualized computing instances and their snapshot hierarchies between object stores with different snapshot implementations, such as redo log and native snapshots, is challenging due to incompatibilities in snapshot management and write operations, leading to failed chain copy operations and incorrect object running points.

Innovation Solution

A methodological approach to migrate VMs involves identifying and migrating disk chains, creating native snapshots, and reverting object running points to ensure compatibility between redo log and native snapshot implementations, allowing successful transfer of snapshot hierarchies across object stores.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If chain copy operations are performed between object stores with different snapshot implementations, then migration of virtual machines is enabled, but the operations fail due to incompatibilities in snapshot management mechanisms

Engineering Contradiction:

Improvemigration compatibilityVSAvoidoperation success rate

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent introduces an intermediary conversion process that translates snapshot hierarchies from the source object store's format (e.g., redo log snapshots with multiple disk chains) into the destination object store's format (e.g., native snapshots with single running point). This intermediary conversion layer resolves the incompatibility between different snapshot implementations, enabling successful migration operations.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the structural parameters of the snapshot hierarchy during migration. It transforms the snapshot representation from one implementation's parameter set (multiple disk chains, different metadata structures) to another implementation's parameter set (single running point, different metadata format), thereby achieving compatibility between heterogeneous object stores.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If disk chains are copied from source to destination object store, then snapshot data is transferred, but the object running point becomes incorrect due to different snapshot implementation mechanisms

Engineering Contradiction:

Improvedata transfer efficiencyVSAvoidobject running point accuracy

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent performs preliminary identification of the object running point in the source snapshot hierarchy before the copy operation. By pre-determining which disk chain and snapshot represent the running point, the system can accurately reconstruct this information in the destination object store after copying, ensuring the running point accuracy is maintained despite the different snapshot implementations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent copies the snapshot hierarchy structure along with metadata that identifies the running point. It reproduces the running point information in the destination object store by creating appropriate snapshot references and updating the running point pointer, thereby maintaining accuracy while enabling efficient data transfer.

Inventive Principle:

Principle #26Copying

3Reliability

If snapshot hierarchies with multiple disk chains are migrated, then complete VM state is transferred, but the complexity of managing different snapshot implementations increases

Engineering Contradiction:

ImproveVM state completenessVSAvoidsnapshot management complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the snapshot hierarchy into individual disk chains for independent processing. By dividing the complex multi-chain structure into manageable segments, the system can process and convert each chain separately according to the destination object store's snapshot implementation requirements, reducing the overall management complexity while preserving complete VM state.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent inverts the traditional migration approach by converting the snapshot hierarchy structure to match the destination object store's implementation rather than requiring the destination to adapt to the source. This inversion simplifies the migration process by applying a standardized conversion methodology that handles the complexity centrally in the conversion logic rather than requiring complex coordination between heterogeneous systems.

Inventive Principle:

Principle #13The other way round (Inversion)

Data Source

PatentUS11782628B2Migration of virtualized computing instance with multiple disk chains in virtualized computing environment

Publication Date: 2023.10.10 VMWARE INC

AI summary

Example methods and systems to perform a migration of a virtualized computing instance and its first snapshot hierarchy from a first object store to a second object store have been disclosed. One example method includes identifying a first disk chain of the first snapshot hierarchy having an object running point, identifying a second disk chain of the first snapshot hierarchy different from the first disk chain, and migrating the second disk chain from the first object store to the second object store to form a first branch of a second snapshot hierarchy in the second object store. After the migrating, the example method includes instructing to take a first native snapshot on the object running point in the second object store, instructing to revert the object running point along the first branch and migrating the first disk chain from the first object store to the second object store.