Heterogeneous SSD Controller Namespace Management

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing computer systems and enterprise servers heavily rely on legacy HDD infrastructure, and there is a need for an improved technique to adapt solid state drives (SSDs) with smaller form factors and heterogeneous formats to be used with HDD infrastructures, while also managing storage capacity according to user-defined parameters.

Innovation Solution

A heterogeneous integrated solid state drive is developed, comprising multiple SSDs of different types, a controller, and an interface. The controller receives user-defined memory parameters such as capacity, quality of service level, I/O operations per second, bandwidth, and latency, and creates namespaces to satisfy these parameters, with the ability to autonomously move data to implement quality of service levels, using protocols like NVMe over Fabric and PCIe.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If solid state drives with smaller form factors and heterogeneous formats are integrated into HDD infrastructures, then storage efficiency and performance are improved, but compatibility and management complexity worsen

Engineering Contradiction:
Improvestorage efficiencyVSAvoidmanagement complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces a controller as an intermediary component that manages multiple SSDs with different form factors and memory types (SLC, MLC, TLC, QLC). The controller abstracts the heterogeneity of the SSDs and presents a unified interface to the host system, enabling compatibility with existing HDD infrastructures while maintaining storage efficiency improvements. This mediator handles the complexity of managing diverse SSD types internally without exposing it to the external system.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The integrated solid state drive system is designed with universal compatibility to work within existing HDD infrastructures while supporting multiple SSD form factors (2.5-inch, M.2) and memory types. The system can adapt to different configurations and present unified storage capacity to the host, making it multi-functional and broadly compatible without requiring infrastructure changes.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Adaptability or versatility

If namespaces are created to satisfy user-defined parameters, then storage management flexibility is improved, but device complexity worsens

Engineering Contradiction:
Improvestorage management flexibilityVSAvoidcontroller complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The controller dynamically creates and manages namespaces based on user-defined parameters such as capacity, quality of service level, I/O operations per second, bandwidth, and latency. The namespace configuration is not static but can be adjusted and optimized based on varying requirements, providing flexibility while the controller handles the complexity of dynamic management internally.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system allows user-defined parameters to be changed and applied to namespace creation and management. Different namespaces can be configured with different parameters (capacity, QoS level, I/O operations, bandwidth, latency) to meet specific application requirements, providing adaptability without requiring changes to the underlying hardware architecture.

Inventive Principle:
Principle #35Parameter changes

3Quantity of substance

If heterogeneous solid state memory devices are integrated, then storage capacity and cost-effectiveness are improved, but performance consistency worsens

Engineering Contradiction:
Improvestorage capacityVSAvoidperformance consistency
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The system integrates different types of solid state memory devices (SLC, MLC, TLC, QLC) each with their own local characteristics and performance profiles. SLC devices provide faster write speeds and higher endurance, while QLC devices provide higher capacity at lower cost. The controller manages these devices with different local qualities, allocating workloads appropriately to maintain overall performance consistency while achieving high storage capacity.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The integrated solid state drive uses a composite architecture combining multiple types of memory devices (SLC, MLC, TLC, QLC) in a single system. This composite approach allows the system to leverage the strengths of each memory type (endurance, speed, capacity, cost) while the controller coordinates them to provide consistent overall performance, achieving both high capacity and performance reliability.

Inventive Principle:
Principle #40Composite materials

Data Source

PatentUS10866912B2Integrated heterogeneous solid state storage drive
Publication Date: 2020.12.15 KIOXIA CORP
  • US10866912B2 patent drawing
  • US10866912B2 patent drawing
  • US10866912B2 patent drawing

AI summary

In one embodiment, a heterogeneous integrated solid state drive includes a plurality of solid state memory devices including at least one solid state memory device of a first type and at least one solid state memory device of a second type, a controller coupled to each of the plurality of solid state memory devices and an interface coupled to the controller. The controller is configured to receive at least one user-defined memory parameter and to create at least one namespace satisfying the at least one user-defined memory parameter in at least one of the plurality of solid state memory devices. In one embodiment, the at least one user-defined memory parameter is one of a group consisting of a capacity, a quality of service level, an assured number of I/O operations per second, a bandwidth, a latency, and an endurance.