Memory Controller Metadata Management for Non-Volatile Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current memory systems face challenges in effectively managing metadata to enhance performance and ensure data integrity, particularly in non-volatile memory devices, where abnormal power-offs can lead to data loss and require efficient restoration methods.
Innovation Solution
A method and system for managing metadata by generating log entry information for changes, storing selected metadata groups and associated log data in non-volatile memory, and using a round-robin scheme to select and store data, ensuring data integrity and efficient restoration upon power-on after an abnormal power-off.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If metadata is stored in buffer memory only, then access speed is improved, but data loss occurs during abnormal power-off
Solution Approach 1:
The patent divides metadata into multiple groups (first metadata group, second metadata group, etc.) and stores them in different locations (buffer memory and non-volatile memory). This segmentation allows the system to maintain fast access for recently modified metadata while ensuring critical metadata is preserved in non-volatile memory, resolving the contradiction between speed and reliability.
Solution Approach 2:
The system performs preliminary actions by storing selected metadata groups in non-volatile memory before abnormal power-off occurs. The memory controller proactively transfers metadata groups to non-volatile memory based on usage patterns and importance, ensuring data integrity is maintained without significantly impacting access speed during normal operation.
2Reliability
If all metadata is stored in non-volatile memory, then data integrity is improved, but access speed deteriorates
Solution Approach 1:
The patent segments metadata into different groups and stores them in different memory locations based on their access frequency and importance. Frequently accessed metadata remains in buffer memory for fast access, while less frequently accessed but critical metadata is stored in non-volatile memory for reliability, thus resolving the contradiction between speed and integrity.
Solution Approach 2:
Different parts of the metadata have different storage qualities - critical metadata is stored in non-volatile memory for reliability, while less critical metadata is stored in buffer memory for speed. This local differentiation allows the system to optimize both reliability and access speed simultaneously.
3Reliability
If metadata groups are selectively stored in non-volatile memory, then data integrity during power-off is improved, but restoration complexity increases
Solution Approach 1:
The system performs preliminary actions by storing metadata groups in non-volatile memory before power-off, and the restoration process leverages this pre-positioned data. During power-on, the system only needs to restore the selected metadata groups that were stored, rather than recovering all metadata, thus reducing restoration complexity while maintaining integrity.
Solution Approach 2:
The patent extracts only the necessary metadata groups for restoration and stores them in non-volatile memory. This extraction approach allows the system to maintain data integrity for critical metadata while simplifying the restoration process by focusing only on the extracted and essential metadata groups rather than all metadata.
4Reliability
If log entry information is stored continuously, then data restoration capability is improved, but memory space consumption increases
Solution Approach 1:
The system applies partial action by storing only the necessary log entry information related to selected metadata groups in non-volatile memory, rather than storing all log entries continuously. This partial storage approach maintains sufficient restoration capability for critical data while reducing overall memory space consumption.
Solution Approach 2:
The patent extracts and stores only the essential log entry information corresponding to selected metadata groups in non-volatile memory. This extraction allows the system to maintain data restoration capability for critical metadata while minimizing the quantity of data stored, thus reducing memory space consumption.
Data Source
AI summary
A method of managing meta data can be provided by generating log entry information including log data in response to changes to meta data that includes a plurality of groups of the meta data. A group of the meta data can be selected from among the plurality of groups of the meta data to provide a selected group of meta data in response to detecting that a number of pieces of the log entry information is equal to or greater than a particular threshold value. The selected group of the meta data and associated log data can be stored in a non-volatile memory device.