Hybrid SSD Data Relocation via Access Pattern Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Information handling systems face increasing demands for higher storage density and faster access times due to larger file sizes and user responsiveness, while also needing to reduce system size, which existing technologies struggle to meet effectively.

Innovation Solution

A hybrid storage device is implemented, utilizing a combination of memory cell types such as quad-level cells (QLC) for higher density and single-level cells (SLC) for quicker access, with data prioritization and migration between these types based on access patterns to optimize storage and access speed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If QLC memory cells are used for storage, then storage density is improved, but access speed deteriorates

Engineering Contradiction:
Improvestorage densityVSAvoidaccess speed
Core Design Contradiction:
Quantity of substanceVSSpeed

Solution Approach 1:

The storage device is segmented into multiple pools with different memory cell types (QLC, TLC, MLC, SLC), each serving different data types. QLC pools handle capacity-intensive cold data while SLC pools handle performance-intensive hot data, resolving the speed-density tradeoff through functional segmentation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different regions of the storage device have different memory cell characteristics optimized for different purposes. QLC regions provide high density for specific data types while SLC regions provide high speed for other data types, allowing local optimization rather than uniform performance across the entire device.

Inventive Principle:
Principle #3Local quality

2Speed

If SLC memory cells are used for storage, then access speed is improved, but storage density deteriorates

Engineering Contradiction:
Improveaccess speedVSAvoidstorage density
Core Design Contradiction:
SpeedVSQuantity of substance

Solution Approach 1:

The storage device is segmented into multiple pools with different memory cell types (QLC, TLC, MLC, SLC), each serving different data types. QLC pools handle capacity-intensive cold data while SLC pools handle performance-intensive hot data, resolving the speed-density tradeoff through functional segmentation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different regions of the storage device have different memory cell characteristics optimized for different purposes. QLC regions provide high density for specific data types while SLC regions provide high speed for other data types, allowing local optimization rather than uniform performance across the entire device.

Inventive Principle:
Principle #3Local quality

3Speed

If data is migrated between memory cell types, then access performance is improved, but device complexity increases

Engineering Contradiction:
Improveaccess performanceVSAvoiddevice complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The storage device autonomously monitors its own performance metrics and automatically migrates data between pools based on detected conditions. The controller self-manages the data migration process without external intervention, using performance degradation detection to trigger automatic data reallocation between different memory cell types.

Inventive Principle:
Principle #25Self-service

4Volume of moving object

If system size is reduced, then portability is improved, but storage capacity deteriorates

Engineering Contradiction:
Improvesystem sizeVSAvoidstorage capacity
Core Design Contradiction:
Volume of moving objectVSQuantity of substance

Solution Approach 1:

The storage device uses a composite architecture combining multiple memory cell types (QLC, TLC, MLC, SLC) in a unified controller. This composite approach allows the device to achieve high storage capacity through QLC while maintaining high performance through SLC, effectively increasing capacity per unit volume without proportionally increasing device size.

Inventive Principle:
Principle #40Composite materials

Data Source

PatentUS20240118829A1Relocation of data in a storage device based on access information
Publication Date: 2024.04.11 DELL PROD LP
  • US20240118829A1 patent drawing
  • US20240118829A1 patent drawing
  • US20240118829A1 patent drawing

AI summary

Aspects of this disclosure improve data availability by decreasing access times to obtain data from the storage device while still providing high memory density. A method may include identifying, by a controller of a non-volatile solid-state storage device, first data that matches a criteria based on access information regarding data stored in a solid state drive as targeted data; writing, by the controller of the non-volatile solid-state storage device, the targeted data from a first portion of the non-volatile solid-state storage device configured as a first type to a second portion of the non-volatile solid-state storage device configured as a second type; and tagging, by the controller of the non-volatile solid-state storage device, the data with a tag based on the access information. Other aspects are also disclosed.