Remote Data Replication for Shared Filesystem Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Concurrently accessed shared filesystems in storage clusters face challenges in failover, as multiple independent data nodes complicate the process, potentially disrupting access when the storage system becomes unavailable.
Innovation Solution
Implementing a Remote Data Replication (RDR) facility between storage clusters to replicate the concurrently accessed shared filesystem and associated filesystems, allowing for seamless failover by mirroring data from a primary storage array to a backup array, ensuring continued access and enhanced availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple independent data nodes are used to provide concurrent access to shared filesystem, then accessibility and parallel access capability are improved, but failover complexity and system reliability are worsened
Solution Approach 1:
The system segments the storage cluster into multiple independent data nodes, each capable of accessing the shared filesystem. This segmentation enables concurrent access while maintaining individual node independence, resolving the contradiction between accessibility and failover reliability by allowing any node to take over in case of failure.
Solution Approach 2:
The patent changes the operational parameters of the storage system by implementing Remote Data Replication (RDR) groups that replicate data across multiple sites. This parameter change enables automatic failover mechanisms, improving reliability while preserving concurrent access capabilities through coordinated parameter management across nodes.
2Loss of time
If Remote Data Replication is implemented to replicate all data nodes and shared filesystem, then failover speed and availability are improved, but data replication complexity and storage resources are worsened
Solution Approach 1:
The system performs preliminary actions by pre-configuring RDR groups and establishing replication facilities before failover events occur. Data is continuously replicated to standby nodes in advance, so when failover is needed, the process is already prepared and can execute immediately, reducing failover time while managing complexity through automation.
Solution Approach 2:
The patent implements copying mechanisms through RDR facilities that create and maintain copies of data across multiple storage nodes and sites. This copying approach enables rapid failover by having pre-existing data copies ready to take over, while the complexity is managed through standardized copying protocols and automated replication management.
Data Source
AI summary
A storage cluster includes a group of data nodes having concurrent access to a shared filesystem. The shared filesystem is assigned to a first TLU of a first storage group. Other filesystems of the data nodes associated with the shared filesystem, such as snapshot copies of the shared filesystems and the root/configuration filesystems of the data nodes are assigned to TLUs of a second set of storage groups. The first storage group and the second set of storage groups are all associated with a Remote Data Replication (RDR) group for the storage cluster. An RDR facility is created between a storage array storing the shared filesystem and a backup storage array. The concurrently accessed shared filesystem and the filesystems of all data nodes accessing shared filesystem are replicated on the RDR facility to the backup storage array to enable failover of the storage cluster between sites.