Metadata Attribute Table for Multi-Type Nonvolatile Memory Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage devices with multiple nonvolatile memory types face inefficiencies in accessing and managing metadata due to the lack of optimized memory selection based on data attributes, leading to suboptimal performance and reliability.
Innovation Solution
A method and storage device that classify metadata based on attributes and accessible memory types, determining the most suitable nonvolatile memory for each metadata type to perform access operations, utilizing a meta data attribute table to direct operations to the optimal memory for read, program, or erase operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If metadata is stored in a single nonvolatile memory type, then device complexity is reduced, but performance and reliability are degraded due to inability to optimize storage based on data attributes
Solution Approach 1:
The patent segments metadata into different categories based on attributes (e.g., frequently accessed metadata, rarely accessed metadata, critical metadata) and assigns each category to appropriate nonvolatile memory types. This segmentation enables optimized access patterns for different metadata while maintaining manageable complexity through systematic classification rules.
Solution Approach 2:
The patent applies local quality by assigning different storage characteristics to different metadata based on their specific attributes. Critical metadata with high reliability requirements are stored in more reliable memory types, while less critical metadata use standard memory types. This localized optimization improves overall system performance without requiring complete rearchitecture.
2Reliability
If multiple nonvolatile memory types are used for metadata storage, then performance and reliability are improved, but device complexity increases due to need for memory selection and management
Solution Approach 1:
The patent performs preliminary action by pre-classifying metadata attributes and pre-determining optimal memory assignments during system initialization or metadata creation. This advance preparation eliminates the need for complex real-time decision-making during metadata access operations, thereby improving reliability while keeping the runtime management simple.
Solution Approach 2:
The patent introduces an intermediary metadata attribute table that acts as a mediator between metadata and nonvolatile memories. This table stores mapping information between metadata identifiers and their assigned memory locations, simplifying the selection process by providing a direct lookup mechanism rather than requiring complex evaluation of multiple memory characteristics during access operations.
3Speed
If metadata access operations are performed without attribute-based classification, then device complexity is reduced, but access speed and performance are degraded
Solution Approach 1:
The patent applies preliminary action by pre-classifying metadata attributes and pre-determining optimal memory assignments during system initialization or metadata creation. This advance preparation eliminates the need for complex real-time decision-making during metadata access operations, thereby improving reliability while keeping the runtime management simple.
Solution Approach 2:
The patent introduces an intermediary metadata attribute table that acts as a mediator between metadata and nonvolatile memories. This table stores mapping information between metadata identifiers and their assigned memory locations, simplifying the selection process by providing a direct lookup mechanism rather than requiring complex evaluation of multiple memory characteristics during access operations.
Data Source
AI summary
A method of accessing data in a storage device including first and second nonvolatile memories of different types is provided. The method includes setting a meta data attribute table by classifying a plurality of meta data based on a plurality of data attributes and accessible memory types, detecting a data attribute of first meta data among the plurality of meta data based on the meta data attribute table in response to receiving a first access request for the first meta data, determining a target memory optimized for the first meta data from among the first and second nonvolatile memories based on the detected data attribute of the first meta data, and performing an access operation on the target memory based on the first meta data. The plurality of meta data are used for controlling an operation of the storage device.