Active/Active Remote Mirroring for Clustered Storage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data replication and mirroring systems face issues with latency in synchronous transfers, data consistency in asynchronous transfers, and active/passive operational behavior, which are inadequate for clustered applications requiring symmetric storage and high availability.

Innovation Solution

An active/active mirroring system is implemented, where both storage devices are accessible for I/O operations and control mirroring operations synchronously with host I/O, allowing for automatic failover and lock conflict resolution, ensuring data consistency and high availability without the need for explicit failover or server mirroring.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If synchronous RDF transfer is used to ensure data consistency between primary and secondary storage devices, then data consistency is improved, but latency is increased due to waiting for acknowledgment

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

Solution Approach 1:

The system performs preliminary actions by having the primary storage device send data to the secondary storage device before receiving acknowledgment. The write operation is considered complete once the secondary device acknowledges receipt, allowing the host to proceed without waiting for full synchronization confirmation, thus reducing latency while maintaining data consistency.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the synchronization behavior based on operational conditions. During normal operation, synchronous transfer ensures consistency, but the system can transition to asynchronous operation when latency becomes problematic, allowing flexible optimization between consistency and performance based on real-time needs.

Inventive Principle:
Principle #15Dynamics

2Loss of time

If asynchronous background copy is used to reduce latency, then latency is improved, but data consistency cannot be guaranteed between primary and secondary storage devices

Engineering Contradiction:
ImprovelatencyVSAvoiddata consistency
Core Design Contradiction:
Loss of timeVSReliability

Solution Approach 1:

The system implements feedback mechanisms where the secondary storage device sends acknowledgment signals back to the primary device to confirm successful data receipt. This feedback loop allows the system to verify data consistency without requiring the host to wait, as the acknowledgment process occurs in the background while providing confirmation that consistency has been maintained.

Inventive Principle:
Principle #23Feedback

3Device complexity

If active/passive operational behavior is used in RDF systems, then device complexity is reduced, but adaptability is decreased for clustered applications requiring symmetric storage

Engineering Contradiction:
Improveoperational behaviorVSAvoidsymmetric storage capability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The system implements multi-functionality by enabling both storage devices to operate in active roles simultaneously. Each device can serve as a primary device for different hosts or data sets, providing universal active/active capability that supports clustered applications and symmetric storage requirements while maintaining manageable complexity through standardized protocols.

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

4Reliability

If synchronous RDF transfer is used over long distances, then data consistency is improved, but transmission delays are increased making the transfer time undesirable

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

Solution Approach 1:

The system performs preliminary data transfer actions over the long-distance connection before the host completes its write operation. By initiating the transfer in advance and using acknowledgment-based completion, the system reduces the perceived transfer time for the host while ensuring data consistency is maintained through the acknowledgment protocol.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS8335899B1Active/active remote synchronous mirroring
Publication Date: 2012.12.18 EMC IP HLDG CO LLC
  • US8335899B1 patent drawing
  • US8335899B1 patent drawing
  • US8335899B1 patent drawing

AI summary

An active/active remote mirroring system, for example an active/active SRDF system, provides for remote mirroring between multiple storage volumes, e.g., a first storage device and a second storage device, in which the storage devices may both be active. A plurality of hosts may be coupled to either the first storage device, the second storage device, or both to conduct read and write operations from and to the storage devices, for example to nearest device. The hosts may be part of an application cluster. The first storage device and the second storage device may be connected via a link, such as an SRDF link, for providing remote mirroring capabilities between the storage volumes. In various embodiments the first and second storage devices may be separate devices, volumes, or portions thereof, and/or may be different portions of the same storage device.