Global Cache Active-Active Storage Replication

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional data protection systems face challenges in providing efficient and asynchronous replication and mirroring across multiple sites, leading to data lag and complexity in transferring primary storage functionality, which can impact disaster recovery and data accessibility.

Innovation Solution

A method and apparatus that utilize a global cache coupled with virtual storage volumes to provide synchronous and asymmetric active-active-active access, enabling asynchronous replication and mirroring across multiple sites, and automatically transferring primary storage functionality between sites to minimize data lag and ensure data consistency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If asynchronous replication is used across multiple sites, then data protection coverage is improved, but data lag between sites increases

Engineering Contradiction:
Improvedata protection coverageVSAvoiddata lag
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system segments the multi-site replication architecture into hierarchical levels: synchronous replication pairs (Site 1-Site 2, Site 1-Site 3) for critical data with zero lag, and asynchronous replication (Site 1-Site 4) for extended coverage. This segmentation allows each segment to operate with optimized latency characteristics while collectively providing comprehensive data protection across all sites.

Inventive Principle:
Principle #1Segmentation

2Reliability

If primary storage functionality is transferred between sites, then disaster recovery capability is improved, but system complexity increases

Engineering Contradiction:
Improvedisaster recovery capabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system implements dynamic role assignment where storage sites can transition between primary and secondary roles based on operational needs. The virtualization layer enables flexible redistribution of primary storage functionality across sites, allowing the system to adapt to failure scenarios and performance requirements without manual reconfiguration, thereby managing complexity through automated dynamic adjustment.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

Each storage site is designed with multi-functionality, capable of operating as primary or secondary site depending on requirements. The virtualization infrastructure provides universal access mechanisms that work regardless of which site holds primary functionality, eliminating the need for site-specific configurations and reducing overall system complexity.

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

3Manufacturing precision

If synchronous mirroring is implemented, then data consistency is improved, but replication speed decreases

Engineering Contradiction:
Improvedata consistencyVSAvoidreplication speed
Core Design Contradiction:
Manufacturing precisionVSSpeed

Solution Approach 1:

The system applies different replication quality characteristics to different data paths based on local requirements: synchronous mirroring is applied to critical site pairs (Site 1-Site 2, Site 1-Site 3) where data consistency is paramount, while asynchronous replication is used for non-critical paths (Site 1-Site 4) where speed is more important. This local quality differentiation resolves the contradiction by matching replication characteristics to specific operational needs.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS9081842B1Synchronous and asymmetric asynchronous active-active-active data access
Publication Date: 2015.07.14 EMC IP HLDG CO LLC
  • US9081842B1 patent drawing
  • US9081842B1 patent drawing
  • US9081842B1 patent drawing

AI summary

In one aspect, a method to provide synchronous and asymmetric active-active-active access to a first storage volume using a global cache coupled to a first, second and third virtual storage volumes includes replicating asynchronously the first storage volume at a first site to a second storage volume at a second site and mirroring the first storage volume with a third storage volume at a third site. Either one of the first storage volume or the third storage volume has primary storage functionality. The method also includes virtualizing the first storage volume with the first virtual storage volume, virtualizing the second storage volume with the second virtual storage volume and virtualizing the third storage volume with the third virtual storage volume.