Storage Node Failover With Log-Based Memory Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face a trade-off between reliability and memory capacity efficiency, as they lose dirty data when all storage control software in a redundancy group fails, leading to impaired reliability and decreased memory capacity efficiency.
Innovation Solution
A storage system with a cluster control unit that disperses storage control units across multiple nodes, synchronizes cache data, and uses logs to restore memory content in alternative nodes upon failure, ensuring high reliability while maintaining memory capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of storage control software in the same redundancy group is increased to reduce the possibility of simultaneous failure, then reliability is improved, but memory capacity efficiency decreases
Solution Approach 1:
The patent creates a copy of the storage node (alternative storage node) when a failure is detected. The alternative storage node replicates the storage control software and memory structure of the failed node, allowing it to take over the failed node's functions. This copying approach enables reliability improvement through redundancy while maintaining memory capacity efficiency because the alternative node uses its own independent memory resources rather than requiring additional memory capacity in the existing nodes.
2Productivity
If storage control software is disposed in the same storage node to efficiently use resources, then resource utilization is improved, but the risk of simultaneous failure increases
Solution Approach 1:
The patent segments the storage control software into multiple redundancy groups, where each group is disposed in different storage nodes. This segmentation ensures that a failure in one storage node does not affect other redundancy groups. The alternative storage node further segments the risk by providing a separate, isolated environment that can be activated without affecting the primary nodes, thus maintaining resource utilization while reducing simultaneous failure risk.
3Reliability
If cache data is synchronized among storage control software in the same redundancy group, then failover capability is improved, but memory capacity efficiency decreases
Solution Approach 1:
The patent introduces a log as an intermediary mechanism for data recovery. Instead of continuously synchronizing cache data among all storage control software (which would consume excessive memory), the system uses logs to record data changes. When a failure occurs, the alternative storage node uses these logs to restore its cache data, achieving failover capability without requiring continuous memory synchronization among all nodes.
Data Source
AI summary
In a storage system including a plurality of storage nodes and a plurality of storage devices that provide storage areas to the storage nodes, a cluster control unit that has detected occurrence of a failure in another storage node requests an external control device to execute a detach process of detaching a storage device allocated to a failure storage node from the failure storage node and an attach process of allocating the storage device to an alternative storage node. Then, the alternative cluster control unit restores, based on a log stored in a storage device allocated to a storage node in which a most recent failure has occurred among the storage nodes belonging to the same redundancy group as a failure storage control unit of the failure storage node, storage content of a memory of the alternative storage node.