Memory Controller Garbage Collection via Data Attribute Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The frequent execution of garbage collection procedures in data storage devices degrades access performance due to the high resource requirements, leading to inefficient data management and processing.

Innovation Solution

A data processing method where the memory controller selects spare memory blocks as destination blocks based on data attributes, moving valid data with the same attribute to the same block, thereby reducing the frequency of garbage collection and optimizing data distribution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If garbage collection procedure is executed frequently to maintain data storage efficiency, then data management effectiveness is improved, but access performance deteriorates due to high resource requirements

Engineering Contradiction:
Improvedata management effectivenessVSAvoidaccess performance
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The patent segments memory blocks into different types (first type for hot data, second type for cold data) and segments the garbage collection process into targeted operations on specific memory block types. This segmentation allows the system to perform garbage collection more selectively and efficiently, reducing the overall resource consumption and frequency of garbage collection operations while maintaining data management effectiveness.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of data classification by introducing attribute-based categorization (hot data vs. cold data) and implements different garbage collection strategies for different data types. This parameter change enables more efficient resource allocation and reduces the frequency of full garbage collection operations, thereby improving access performance while maintaining data management effectiveness.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If valid data is moved to different destination blocks without attribute consideration, then garbage collection completes the migration task, but resource usage increases and performance decreases

Engineering Contradiction:
Improvegarbage collection efficiencyVSAvoidresource usage
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent applies local quality by determining attributes of valid data and moving data with the same attribute to the same destination memory block type. This localized approach optimizes the garbage collection process by concentrating similar data together, reducing the scope of future garbage collection operations, and minimizing resource usage while maintaining high garbage collection efficiency.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent performs preliminary classification of valid data by attribute during the garbage collection process, determining which destination memory block type each data item should belong to before migration. This preliminary action organizes data in a way that reduces future garbage collection needs and optimizes resource usage by preventing scattered data distribution.

Inventive Principle:
Principle #10Preliminary action

3Stability of the object's composition

If garbage collection is performed frequently to maintain storage efficiency, then data organization is improved, but access speed deteriorates due to resource consumption

Engineering Contradiction:
Improvedata organizationVSAvoidaccess speed
Core Design Contradiction:
Stability of the object's compositionVSSpeed

Solution Approach 1:

The patent segments data into hot and cold categories and segments memory blocks accordingly, allowing garbage collection to operate more efficiently on specific segments rather than the entire memory space. This segmentation maintains data organization stability while reducing the frequency and resource consumption of garbage collection operations, thereby preserving access speed.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the organizational parameter by introducing attribute-based data classification and implements differentiated garbage collection strategies for different data types. This parameter change improves data organization stability through systematic categorization while reducing access speed degradation by minimizing the frequency of full garbage collection operations.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11204864B2Data storage devices and data processing methods for improving the accessing performance of the data storage devices
Publication Date: 2021.12.21 SILICON MOTION INC
  • US11204864B2 patent drawing
  • US11204864B2 patent drawing
  • US11204864B2 patent drawing

AI summary

A data storage device includes a memory device and a memory controller. The memory device includes multiple memory blocks. The memory controller is coupled to the memory device and is configured to access the memory device. In a garbage collection procedure, the memory controller is configured to select multiple spare memory blocks as destination memory blocks and move valid data stored in at least one source memory block into the destination memory blocks. In the garbage collection procedure, the memory controller is further configured to determine an attribute of each valid data and determine which destination memory block to move the valid data into according to the corresponding attribute. Valid data having the same attribute is moved to the same destination memory block.