Storage Node Cache Segmentation for Asynchronous Copy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In storage systems with multiple nodes, the high frequency of writing data from cache to non-volatile media in asynchronous remote copies can lead to performance deterioration and data loss due to power disconnections.
Innovation Solution
Implementing a system where each node determines, based on data type, whether redundancy and non-volatilization are required for cache segments, thereby reducing the frequency of writing to non-volatile media.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is frequently written from cache to non-volatile media in asynchronous remote copy, then data redundancy and protection against power disconnection are improved, but storage system performance deteriorates due to high write frequency
Solution Approach 1:
The invention divides the cache into multiple segments and applies different protection strategies to each segment based on data type. Not all cache data is treated equally - critical data is protected with non-volatile storage while less critical data remains in volatile memory, reducing unnecessary write operations and improving overall performance.
Solution Approach 2:
Different cache segments have different protection characteristics. The system determines locally for each segment whether redundancy and non-volatilization are required based on data type, applying protection only where necessary rather than uniformly across the entire cache, thus optimizing the balance between reliability and performance.
2Reliability
If all cache data is made redundant and written to non-volatile media, then data loss prevention is improved, but the frequency of writes to non-volatile media increases causing performance deterioration
Solution Approach 1:
Instead of applying full protection (redundancy + non-volatilization) to all cache data, the system applies partial protection only to the extent necessary for each data type. This avoids excessive protection actions that would waste time on unnecessary write operations while maintaining adequate protection where needed.
Solution Approach 2:
The system changes the protection parameters (redundancy flag and non-volatile flag) for different cache segments based on data type characteristics. By adjusting these parameters dynamically, the system optimizes the balance between data loss prevention and write operation time for each segment.
Data Source
AI summary
In each of a plurality of nodes constituting a storage system, when there is data to be stored in a segment allocated from a cache, the node determines, for the segment, whether redundancy (redundant data of the data in the segment is transferred to another node) and whether non-volatilization (data in the segment is stored in a non-volatile area) are required according to a type of the data, and controls, based on a result of the determination, whether to perform the redundancy of data in the segment and whether to perform the non-volatilization of data in the segment.


