NVM Storage Management via Image Feature Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile memory (NVM) devices face issues of short lifespan due to frequent write operations causing physical block damage and excessive energy consumption during write operations, which leads to wear and energy inefficiency.
Innovation Solution
A storage space management method that determines an idle block with an image feature most similar to the data to be written, using Local Binary Pattern (LBP) eigenvectors and B+ trees, to minimize energy consumption and extend NVM lifespan by optimizing write operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is written into NVM using conventional storage space allocation methods, then write operations can be performed quickly, but the NVM suffers from excessive wear and high energy consumption
Solution Approach 1:
The system pre-calculates and stores image features (LBP eigenvectors) of idle blocks before write operations occur. When data needs to be written, the pre-computed features enable rapid similarity comparison without real-time computation, maintaining high write speed while enabling wear reduction through intelligent block selection
Solution Approach 2:
The patent transforms the physical block selection problem into a feature space comparison problem by converting data blocks and idle blocks into image features. This parameter transformation allows the system to select blocks based on feature similarity rather than arbitrary allocation, reducing wear by writing similar data to blocks with similar characteristics
2Productivity
If conventional storage allocation is used, then storage space can be efficiently managed, but energy consumption during write operations becomes excessively high
Solution Approach 1:
Image features of all idle blocks are calculated and stored in advance in a feature set. This preliminary computation eliminates the need for real-time feature extraction during write operations, maintaining storage management efficiency while significantly reducing the energy required at the moment of data writing
Solution Approach 2:
Instead of physically examining or repeatedly accessing idle blocks during write operations, the system creates and stores digital copies of block characteristics (image features). These feature copies enable rapid comparison and selection without physical block access, reducing energy consumption while maintaining efficient space management
3Productivity
If frequent write operations are performed to manage storage space, then storage utilization is improved, but physical blocks in NVM are damaged and become invalid
Solution Approach 1:
The patent changes the selection criterion from arbitrary or simple metadata-based allocation to image feature similarity comparison. By transforming blocks into feature representations and selecting based on Euclidean distance in feature space, the system reduces unnecessary writes to blocks that would require extensive data rewriting, thereby reducing wear and physical damage
Solution Approach 2:
The patent converts the potentially harmful effect of frequent writes into a benefit by using image feature analysis to identify blocks that are most suitable for receiving new data. By selecting blocks whose existing content is similar to the incoming data, the system minimizes the number of write operations needed and reduces physical block damage while maintaining high storage utilization
Data Source
AI summary
A storage space management method and apparatus, where the method and apparatus are applied to a non-volatile memory (NVM). In a feature set that includes M image features of M idle blocks in storage space of the NVM, an idle block whose image feature is highly similar to an image feature of data to be written into the NVM is determined such that the data is written into the idle block. In this way, wear and energy consumption problems are considered during storage space allocation, and a write operation of an idle block in storage space of an NVM can consume less energy, thereby extending a life span of the NVM and reducing write operation energy consumption.