Synchronous Replication Cutover Engine for Storage Migration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing storage migration processes are disruptive to clients, causing high latency and potential downtime due to asynchronous data replication methods that block client I/O, leading to application failures and operational issues during cutover.

Innovation Solution

Implementing a synchronous replication process in conjunction with asynchronous replication to maintain low client latency, ensuring that client I/O is quiesced for a shorter duration by synchronizing storage objects before cutover, allowing for non-disruptive migration of storage virtual machines across clusters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If asynchronous replication is used for storage migration, then data can be replicated continuously without blocking client I/O, but the cutover process causes high latency and client disruption

Engineering Contradiction:
Improveclient I/O throughputVSAvoidcutover latency
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by quiescing the storage object before cutover to ensure data consistency, and by pre-synchronizing data during the asynchronous replication phase. This prepares the system in advance to minimize disruption during the actual cutover event.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses periodic asynchronous replication to continuously sync data from the source storage object to the destination, checking for data changes at regular intervals. This periodic synchronization allows the system to maintain low latency by only quiescing when necessary, rather than continuously blocking I/O.

Inventive Principle:
Principle #19Periodic action

2Reliability

If synchronous replication is used to ensure data consistency before cutover, then data integrity is improved, but client I/O latency increases during the synchronization process

Engineering Contradiction:
Improvedata consistencyVSAvoidsynchronization time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Synchronous replication is performed as a preliminary action before cutover to ensure data consistency between source and destination storage objects. This preliminary synchronization guarantees that the destination is ready to receive client I/O without data loss.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically switches between asynchronous and synchronous replication modes. Asynchronous replication is used during normal operation to minimize latency, while synchronous replication is activated temporarily before cutover to ensure consistency, and then switched back to asynchronous after cutover.

Inventive Principle:
Principle #15Dynamics

3Reliability

If client I/O is quiesced during cutover to ensure data consistency, then data integrity is improved, but application performance and availability deteriorate

Engineering Contradiction:
Improvedata integrityVSAvoidapplication throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system quiesces the storage object as a preliminary action immediately before cutover to ensure data consistency, rather than maintaining continuous quiescing. This brief quiescing period is sufficient to guarantee integrity while minimizing impact on application performance.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system maintains continuity of useful action by performing asynchronous replication continuously in the background, so that data synchronization ongoing during normal operation. This eliminates the need for extended quiescing periods, as the destination is already synchronized when cutover occurs.

Inventive Principle:
Principle #20Continuity of useful action

4Productivity

If multiple volumes are migrated simultaneously to improve efficiency, then productivity is improved, but the complexity of managing consistency across volumes increases

Engineering Contradiction:
Improvemigration throughputVSAvoidcutover management complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system segments the migration process into independent volume-level operations, where each volume can be replicated and cut over separately. This segmentation allows multiple volumes to be migrated simultaneously while maintaining manageable complexity through consistent per-volume procedures.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system implements a universal cutover mechanism that can be applied to multiple volumes using the same process and logic. This multi-functional approach handles consistency management across different volumes through a standardized method, reducing the complexity that would arise from volume-specific procedures.

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

Data Source

PatentEP3830681B1Synchronous replication based cutover engine
Publication Date: 2024.06.19 NETAPP INC
  • EP3830681B1 patent drawingFigure 1
  • EP3830681B1 patent drawingFigure 2
  • EP3830681B1 patent drawingFigure 3

AI summary

Techniques are provided for synchronous replication based cutover. An asynchronous replication process is executed to perform asynchronous incremental transfers of data of a storage object from a first computing environment to a replicated storage object at a second computing environment until a cutover criteria is met. A synchronous replication process is executed to synchronously replicate operations, targeting the storage object, to the replicated storage object based upon the cutover criteria being met. A cutover is performed to direct operations from targeting the storage object to targeting the replicated storage object based upon the synchronous replication process reaching a steady state of synchronous replication for sub-objects of the storage object, where operations are committed to both the storage object and the replicated storage object.