Non-Volatile Memory Storage Mode Selection Logic

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods of storing data in solid-state drives are inefficient as they do not consider the type of data being stored, leading to increased storage operations and reduced speed.

Innovation Solution

The implementation of selection logic that communicates with the file system and logical-to-physical translation layer to analyze data attributes and storage cell usage, selecting an appropriate storage mode and allocating cells for efficient data storage in non-volatile memory systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If conventional data storage methods are used without considering data type, then the storage system is simple to operate, but the storage speed is reduced and extra storage operations are required

Engineering Contradiction:
Improvestorage speedVSAvoidstorage management complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent segments data into different types (sequential access data and random access data) and applies different storage modes for each type. The file system divides data chunks based on access patterns, storing sequential data contiguously and random data distributedly, thereby optimizing storage speed for each data category without requiring complex manual management.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic storage management where the storage mode is not fixed but adapts based on data characteristics. The system dynamically determines whether to use contiguous or distributed storage by analyzing data access patterns and file attributes, allowing the storage strategy to change flexibly according to actual usage requirements.

Inventive Principle:
Principle #15Dynamics

2Speed

If data is stored in contiguous locations, then sequential access speed is improved, but random access requires more translation operations

Engineering Contradiction:
Improvesequential access speedVSAvoidtranslation overhead time
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

The patent applies different storage qualities to different data types: sequential access data is stored with high contiguity (local quality optimized for sequential access), while random access data is stored with distributed locations (local quality optimized for random access). This localized optimization allows each data type to benefit from its appropriate storage pattern without affecting others.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the storage parameter (contiguity) based on data characteristics. By detecting whether data exhibits sequential or random access patterns, the system adjusts the storage parameter accordingly - using contiguous storage for sequential data and distributed storage for random data, thereby optimizing both access speed and translation efficiency.

Inventive Principle:
Principle #35Parameter changes

3Adaptability or versatility

If data chunks are stored at different locations to accommodate free storage cells, then storage flexibility is improved, but the number of storage operations increases

Engineering Contradiction:
Improvestorage flexibilityVSAvoidstorage operation efficiency
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent performs preliminary classification of data into sequential or random access categories before storage. By determining the access pattern in advance, the system can pre-select the appropriate storage mode, avoiding the need for multiple corrective storage operations later. This preliminary action reduces the total number of storage operations while maintaining flexibility.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent extracts and separates sequential access data from random access data, handling each type independently with its own optimized storage strategy. By taking out sequential data and storing it contiguously, the system reduces the number of scattered storage operations that would otherwise be required, thereby improving overall storage efficiency.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS9025380B1Management of data storage in a non-volatile memory system
Publication Date: 2015.05.05 SK HYNIX NAND PRODUCT SOLUTIONS CORP
  • US9025380B1 patent drawing
  • US9025380B1 patent drawing
  • US9025380B1 patent drawing

AI summary

According to one configuration, selection logic analyzes received attributes of particular data to be stored in a non-volatile memory system. Depending on present usage of storage cells in the non-volatile memory system and the attributes of the particular data to be stored, the selection logic selects a mode amongst multiple possible storage modes in which to store the particular data in the non-volatile memory system. Thereafter, the selection logic initiates storage of the particular data in the non-volatile memory system in accordance with the selected storage mode.