Slave Storage Array Buffering for Seamless Disaster Recovery Failover

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Previous data storage systems face performance and resource issues when handling local storage array failures, requiring host machine restarts or extensive data replication, which are time-consuming and resource-intensive.

Innovation Solution

Implementing a system with a master and slave storage array configuration, where write operations are collected and replicated synchronously, allowing seamless transition to the slave array in case of master array failure, eliminating the need for host machine restarts and reducing the burden of tracking un-conveyed data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If synchronous replication is implemented between master and slave storage arrays, then disaster recovery reliability is improved, but system complexity increases

Engineering Contradiction:

Improvedisaster recovery protectionVSAvoidstorage array configuration

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the storage array functionality into master and slave roles, with each array capable of independently performing both master and slave operations. This segmentation allows the system to maintain simplified individual array designs while achieving complex disaster recovery capabilities through their interaction.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The slave storage array preliminarily collects and buffers write operations locally before they need to be replicated to the remote array. This preliminary action ensures that when failover occurs, the slave array is already prepared with the necessary data, eliminating the need for host machine restarts and reducing recovery time.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If the host machine restarts to establish access to backup storage array, then disaster recovery is achieved, but productivity is reduced due to service interruption

Engineering Contradiction:

Improvefault toleranceVSAvoidhost machine availability

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The slave storage array acts as an intermediary between the host machine and the remote backup array. It buffers write operations locally and can seamlessly take over as the master array, providing continuous access to hosts without requiring them to restart or re-establish connections.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If host process tracks un-conveyed data for disaster recovery, then data integrity is improved, but use of energy and computational resources increases

Engineering Contradiction:

Improvedata integrityVSAvoidhost machine resources

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent extracts the burden of tracking and buffering un-conveyed write operations from the host machine and places it on the slave storage array. The slave array maintains its own local buffer and can independently manage the replication state, freeing the host from resource-intensive tracking tasks while ensuring data integrity.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The slave storage array serves itself by independently collecting, buffering, and managing write operations locally. It autonomously tracks the state of replicated data and can independently initiate failover procedures, eliminating the need for host processes to perform resource-intensive tracking and management tasks.

Inventive Principle:

Principle #25Self-service

4Reliability

If complete copying of storage object is performed from non-failing local array to remote array, then disaster recovery protection is established, but loss of time and network resources increases

Engineering Contradiction:

Improvedisaster recovery protectionVSAvoiddata replication time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The slave storage array preliminarily collects and buffers write operations locally in real-time as they occur. This continuous preliminary action ensures that the slave array maintains an up-to-date copy of the data, so when failover is needed, the data is already prepared and can be immediately replicated to the remote array without requiring a complete copy operation.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of performing a complete copy of the entire storage object when failover is needed, the system uses the partially replicated data already buffered in the slave array. This partial action (using only the necessary portion of data that has already been collected) dramatically reduces the time and network resources required for disaster recovery establishment.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS10146648B1Preserving disaster recovery protection for a data storage object

Publication Date: 2018.12.04 EMC IP HLDG CO LLC

AI summary

Replication of a storage object to a remote storage array is seamlessly maintained by i) while write operations directed to the storage object are being collected into a set of write cycles that are stored in a master storage array, also collecting the write operations directed to the storage volume into a set of write cycles stored in a slave storage array, and ii) in response to detecting a failure of the master storage array, and while the host computer continues to access the storage object using a copy of the storage object located in the slave storage array, transmitting the write cycles stored in the slave storage array to the remote storage array, so that the remote storage array can apply the write operations in the write cycles it receives from the slave storage array on the copy of the storage object contained in the remote storage array.