Common Snapshot Consistency Point Handling for Low-Latency Replication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Synchronous replication of datasets between primary and secondary storage systems can lead to latency spikes during common snapshot creation due to I/O operation fencing, which can disrupt business operations and exceed network protocol timeouts.

Innovation Solution

Optimizations such as batch processing of synchronous replicated datasets, coalescing consistency points, and asynchronous CP handling are employed to minimize I/O latency during common snapshot creation, utilizing multithreading and maintaining file system consistency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If synchronous replication is performed between primary and secondary storage systems, then data integrity and consistency are improved, but I/O latency increases during common snapshot creation

Engineering Contradiction:
Improvedata integrityVSAvoidI/O latency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the common snapshot creation process into multiple independent consistency point operations, one for each replicated dataset. Instead of creating a single global consistency point that blocks all I/O operations, the system creates individual consistency points for each dataset that can be processed independently and in parallel, reducing the overall I/O latency while maintaining data integrity across the replicated systems.

Inventive Principle:
Principle #1Segmentation

2Reliability

If common snapshot creation is performed across multiple replicated datasets, then data consistency across sites is improved, but I/O operation resumption time increases

Engineering Contradiction:
Improvedata consistencyVSAvoidresumption time
Core Design Contradiction:
ReliabilityVSDuration of action of moving object

Solution Approach 1:

The patent implements preliminary actions by pre-establishing consistency points for each replicated dataset before initiating the common snapshot creation process. This allows the system to have consistency checkpoints ready in advance, so when snapshot creation is needed, the system can quickly reference and utilize these pre-positioned consistency points rather than creating them from scratch during the snapshot operation, thereby reducing resumption time.

Inventive Principle:
Principle #10Preliminary action

3Manufacturing precision

If I/O operations are fenced during common snapshot creation, then snapshot consistency is improved, but business operation continuity deteriorates

Engineering Contradiction:
Improvesnapshot consistencyVSAvoidoperation continuity
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies local quality by implementing fine-grained fencing at the dataset level rather than system-wide fencing. When a common snapshot is being created, only the specific datasets that are being snapshotted have their I/O operations fenced, while other datasets and storage operations continue to function normally. This localized approach maintains snapshot consistency for the targeted datasets while preserving business operation continuity for the rest of the system.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS20250355773A1Methods and systems to reduce latency of input/output (i/o) operations based on consistency point optimizations during creation of common snapshots for synchronous replicated datasets of a primary copy of data at a primary storage system to a mirror copy of the data at a cross-site secondary storage system
Publication Date: 2025.11.20 NETAPP INC
  • US20250355773A1 patent drawing
  • US20250355773A1 patent drawing
  • US20250355773A1 patent drawing

AI summary

Multi-site distributed storage systems and computer-implemented methods are described for improving a resumption time of input/output (I/O) operations during a common snapshotprocedure for storage objects. A computer-implemented method includes initiating a snap create handler operation for a storage object of a batch of storage objects having a plurality of replicated datasets with each replicated dataset having a synchronous replication relationship between at least one storage object of the first storage node and at least one replicated storage object of the second storage node, determining whether a consistency point is currently in progress or not, and providing a hint to accelerate a currently in progress consistency point when the consistency point is currently in progress.