Cascading Data Replication Across Storage Nodes for Disaster Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face inefficiencies in data management and protection, particularly in handling flash storage systems where operations are often performed by both higher and lower level processes, leading to unnecessary write operations and reduced reliability.
Innovation Solution
A direct-mapped flash storage system where operations are initiated and controlled solely by the operating system, eliminating redundant lower level processes and enhancing reliability by direct addressing of data blocks without intermediate storage controller intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If both higher level and lower level processes perform operations on flash storage systems, then data management functionality is enhanced, but unnecessary write operations increase and reliability decreases
Solution Approach 1:
The patent extracts and eliminates the lower level storage controller intervention from the data management process. The storage controller is removed from participating in data management decisions, allowing only the operating system to perform operations. This extraction resolves the contradiction by removing the source of redundant write operations while preserving necessary functionality through OS-level management.
Solution Approach 2:
The patent introduces a direct communication path between the operating system and flash storage devices, eliminating the storage controller as an intermediary. This direct addressing mechanism allows the OS to manage data blocks without intermediate intervention, reducing unnecessary write operations while maintaining full data management capability through the OS.
2Reliability
If storage controllers intervene in data block operations, then data protection is provided, but data management efficiency decreases
Solution Approach 1:
The patent extracts the storage controller from the data management pathway, removing it as an intermediate component. Data blocks are addressed directly by the operating system without storage controller intervention. Data protection is maintained through OS-level error handling and redundancy mechanisms, while efficiency improves by eliminating the controller bottleneck.
Solution Approach 2:
The flash storage system is designed to be self-managing at the device level, with each flash storage device capable of directly receiving and executing data management operations from the operating system. This self-service capability eliminates dependency on storage controller mediation, improving efficiency while maintaining data integrity through built-in device-level error detection and correction.
Data Source
AI summary
Cascading data replication for disaster recovery, including: establishing a cascading replication relationship between a plurality of storage nodes comprising a source endpoint node, a target endpoint node, and one or more intermediary nodes, wherein the cascading replication relationship comprises a chain of replication relationships between pairs of storage nodes from the source endpoint node to the target endpoint node via the one or more intermediary nodes; and replicating, based on the cascading replication relationship, a dataset from the source endpoint node to the target endpoint node via the one or more intermediary nodes.