Nonvolatile Memory Copy History for Hot-Cold Data Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The mixing of hot and cold data in a nonvolatile memory block leads to a decrease in the efficiency of garbage collection, as it is difficult to determine the type of data without additional information from the host device, resulting in inefficient resource management.
Innovation Solution
A memory system that includes a controller to write data and history information to a block, where the history information is based on the number of times the data has been copied, allowing the system to distinguish between hot and cold data by increasing the history information value each time data is copied, thereby optimizing garbage collection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If hot data and cold data are mixed together in a same block of the nonvolatile memory, then the storage capacity is utilized, but the efficiency of garbage collection decreases
Solution Approach 1:
The patent segments data into hot data and cold data based on write frequency characteristics, and manages them separately through different garbage collection strategies. Hot data blocks undergo frequent GC operations while cold data blocks undergo less frequent GC, resolving the contradiction by dividing the monolithic GC process into targeted segments.
Solution Approach 2:
The patent changes the parameter of garbage collection frequency based on data type. Instead of uniform GC across all blocks, hot data blocks receive higher GC frequency and cold data blocks receive lower GC frequency, optimizing overall GC efficiency while accounting for data characteristics.
2Measurement precision
If additional information is requested from the host device to determine data type, then data classification accuracy improves, but communication overhead increases
Solution Approach 1:
The memory device performs self-service by autonomously determining whether data is hot or cold based on its own monitoring of write operations. The controller tracks write frequencies and automatically classifies data without requiring host device input, eliminating communication overhead while maintaining classification accuracy.
Solution Approach 2:
The system implements feedback mechanisms where the controller continuously monitors write operations to data blocks and uses this feedback to dynamically adjust GC frequency. The write frequency information feeds back into the GC decision-making process, enabling automatic data type identification and adaptive GC optimization.
Data Source
AI summary
According to one embodiment, a memory system includes a controller configured to write data and history information to one of a plurality of blocks. In copying first data that is valid data stored in a copy source block, to a copy destination block, the controller reads the first data and first history information from the copy source block, the first history information being history information corresponding to the first data, and writes the read first data and second history information to the copy destination block. The second history information is based on a value of the read first history information.