Storage System Cache Control for Battery Failure Data Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage systems with cyclically duplicated caches, an abnormal state of a control apparatus's battery increases the risk of data loss not only for the local cache but also for the mirror cache in another control apparatus, compromising the safety of data across the system.
Innovation Solution
The storage system includes a first and second control apparatus with cache regions and power supplies, where each apparatus switches from a write-back method to a write-through method upon detecting an abnormal battery state, ensuring data safety by synchronizing local cache data with memory devices and reducing the risk of data loss across the system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cyclically duplicated cache configuration is used to improve data safety, then data redundancy is improved, but the risk of data loss in another control apparatus increases when a battery abnormality occurs
Solution Approach 1:
The system segments the cache protection mechanism by control apparatus, allowing independent write control decisions for each control apparatus based on its own battery status. This segmentation prevents a battery abnormality in one control apparatus from affecting the write control and data safety of other control apparatuses, thus resolving the contradiction between data redundancy and data loss risk.
2Productivity
If write-back method is used to improve write performance, then productivity is improved, but data safety is worsened when battery abnormality occurs
Solution Approach 1:
The system dynamically switches between write-back and write-through methods based on real-time battery status monitoring. When battery abnormality is detected, the system transitions from write-back to write-through to ensure data safety. This dynamic adaptation allows the system to maintain high write performance under normal conditions while ensuring data safety when battery issues occur, resolving the contradiction between productivity and reliability.
3Reliability
If write-through method is used to improve data safety, then reliability is improved, but write performance is worsened
Solution Approach 1:
The system employs periodic battery status monitoring and switches write control methods accordingly. Under normal conditions, the system operates in write-back mode for high performance. When battery abnormality is detected through periodic monitoring, it switches to write-through mode to ensure data safety. This periodic checking and conditional switching resolves the contradiction by using write-through only when necessary rather than continuously.
Data Source
AI summary
A first control apparatus includes a first memory unit including a local cache, a first power supply that supplies electric power to the first memory unit, and a control unit. The control unit controls a write into a memory device by a write-back method, using the local cache. The control unit mirrors data of the local cache in a mirror cache of a second control apparatus. The control unit determines whether the mirror cache is included in a second memory unit that receives electric power from a second power supply of the second control apparatus, upon detecting an abnormal state of a battery for supplying electric power to the second memory unit in case of power outage of the second power supply. The second memory unit switches write control for the memory device to a write-through method, when the second memory unit includes the mirror cache.