Synchronous Replication Resynchronization Using In-Progress Write Maps

Solution Overview

Problem

Existing data resynchronization techniques for stretched volumes in synchronous replication configurations are inefficient due to the periodic creation and deletion of recovery snapshots, leading to increased resource consumption and latency.

Innovation Solution

Persistent tracking of in-progress writes at a designated site, with snapshots taken only upon recovery from replication failures, allowing efficient resynchronization by leveraging these writes and snapshots to minimize data transfer and reduce latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If periodic recovery snapshots are created and deleted for resynchronization, then data can be recovered between sites, but resource consumption increases and latency increases

Engineering Contradiction:

Improvedata recovery capabilityVSAvoidresource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary tracking of in-progress write requests at the designated site during normal operation, maintaining a map of these requests before failures occur. This preliminary action eliminates the need for periodic snapshots by preparing the necessary synchronization information in advance, reducing resource consumption while maintaining recovery capability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of creating full periodic snapshots of data, the system copies only the essential metadata about in-progress write requests (the map) to enable resynchronization. This selective copying approach maintains data recovery capability while dramatically reducing the resource consumption associated with full snapshots.

Inventive Principle:

Principle #26Copying

2Reliability

If periodic recovery snapshots are created and deleted for resynchronization, then data can be recovered between sites, but latency increases

Engineering Contradiction:

Improvedata recovery capabilityVSAvoidresynchronization latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary tracking of in-progress write requests at the designated site during normal operation, maintaining a map of these requests before failures occur. This preliminary action eliminates the need for periodic snapshots by preparing the necessary synchronization information in advance, reducing resource consumption while maintaining recovery capability.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of creating full periodic snapshots of data, the system copies only the essential metadata about in-progress write requests (the map) to enable resynchronization. This selective copying approach maintains data recovery capability while dramatically reducing the resource consumption associated with full snapshots.

Inventive Principle:

Principle #26Copying

3Reliability

If synchronous replication configuration is maintained across multiple sites, then data consistency is ensured, but system complexity increases during failure recovery

Engineering Contradiction:

Improvedata consistencyVSAvoidfailure recovery complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system extracts and isolates the complexity of failure recovery by maintaining a dedicated map of in-progress write requests at the designated site. This extracted information structure separates the recovery process from the normal synchronous replication operations, simplifying the overall system architecture while ensuring data consistency across sites.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The map of in-progress write requests acts as an intermediary data structure between the designated site and other sites during failure recovery. This intermediary enables coordinated resynchronization by providing a clear, structured view of pending operations, reducing the complexity of maintaining consistency across multiple sites during failure scenarios.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12554430B2Techniques for optimized data resynchronization between replication sites

Publication Date: 2026.02.17 DELL PROD LP

AI summary

Techniques for resynchronizing storage resources of two sites configured for synchronous replication can include tracking in-progress write requests in a map. Responsive to a site failure or a replication link failure resulting in failure or fracture of the synchronous replication, processing can be performed to resynchronize impacted resources of the two sites configured for synchronous replication. The processing can use the in-progress write requests in combination with resource snapshots and snapshot differences to resynchronize the impacted resource of the two sites. In at least one embodiment, the synchronous replication configuration can include active paths between both sites and a host. In at least one embodiment, the synchronous replication configuration can include an active path between the host and one site, and a passive path between the host and the second site. The synchronous replication can be bi-directional or two-way synchronous replication between the two sites.