Storage Mirror Resynchronization via Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face challenges in efficiently resynchronizing mirror resources, particularly when disaster recovery (DR) nodes become non-responsive, leading to performance issues and the need for additional processing resources, which can result in system downtime and communication losses.
Innovation Solution
A management processing system that controls storage systems to enable multiple mirroring resources with HA and DR functionality at different locations, allowing for transparent resynchronization of nodes without disrupting client activity or affecting other mirror resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a DR node becomes non-responsive and requires resynchronization, then data consistency is restored, but client performance suffers due to system downtime
Solution Approach 1:
The system divides the resynchronization process into segments by allowing it to occur independently on individual mirror resources without affecting other mirrors. The management processing system identifies which specific mirror resource needs resynchronization and isolates that operation, enabling other mirrors to continue serving client requests normally.
Solution Approach 2:
The management processing system acts as an intermediary that coordinates between the primary storage resource, multiple mirror resources, and client devices. It monitors the health of mirror resources, detects when resynchronization is needed, and orchestrates the resynchronization process to minimize impact on client operations.
2Reliability
If traditional resynchronization methods are used, then mirror resources are brought back online, but additional processing resources are required and system complexity increases
Solution Approach 1:
The mirror resource that becomes non-responsive performs self-service resynchronization by automatically detecting its own failure state and initiating the resynchronization process with the primary storage resource without requiring manual intervention or complex external coordination.
Solution Approach 2:
The management processing system performs multiple functions including monitoring mirror resource health, detecting failures, initiating resynchronization, and coordinating between multiple mirrors. This multi-functional approach consolidates what would otherwise require separate specialized components.
3Reliability
If multiple mirror locations are implemented, then disaster recovery capability is improved, but the likelihood of system downtime and communication losses doubles
Solution Approach 1:
The system segments the mirror resources into independent, separately-managed units that can fail and be resynchronized independently. When one mirror experiences downtime, the other mirrors continue to operate normally, preventing cascading failures and reducing overall system downtime frequency.
Solution Approach 2:
While one mirror resource undergoes resynchronization, the management processing system ensures that other mirror resources continue to provide useful actions by serving client requests. This continuity of service from alternative mirrors minimizes overall system downtime.
Data Source
AI summary
Systems and methods which provide for managing multiple mirror resources in a storage distribution network are provided. In some embodiments, a system provides for both high availability and disaster recovery functionality at different mirroring locations. Other embodiments may provide for multiple high availability and/or multiple disaster recovery mirror resources. These mirror resources are operated in a heterogeneous manner in the sense that each have its own transport, protocol, and the like, but are configured function cooperatively or as a single mirror with respect to mirroring a primary node. Embodiments may provide for the mirroring and resynchronization of mirrored resources in the event of a communication loss with a particular resource without ceasing the mirroring operations to other resources.