Chiplet Storage Architecture Segmentation for Protocol Adaptability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data storage systems face limitations in flexibility and compatibility with varying host devices and memory media standards, as they are often designed for specific standards, making them inflexible and unable to adapt to changes in either host devices or memory devices.

Innovation Solution

A storage architecture that includes a front-end chip for interfacing with host devices, back-end chips for interfacing with memory devices, and an input/output chip for communication between them, allowing for separate replacement and updates of each component to support different standards and protocols, enabling adaptability to changes in host and memory device standards.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a storage architecture is configured as a monolithic integrated circuit or system-on-chip, then the structure is simple and manufacturing is easier, but the adaptability to different host devices and memory media standards is limited

Engineering Contradiction:
Improveadaptability to host devices and memory media standardsVSAvoidstorage architecture structure
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The storage architecture is divided into separate functional modules: a first chip for host device interfacing, a second chip for memory device interfacing, and an input/output chip for data transfer. This segmentation allows each chip to be optimized for specific standards and enables independent replacement or upgrade of individual components without affecting the entire system, thereby improving adaptability while managing complexity through modular design

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If a storage architecture is designed for specific host devices and memory media standards, then the design is simplified and manufacturing is easier, but the flexibility to support recent system features and future standards is reduced

Engineering Contradiction:
Improveflexibility to support system features and standardsVSAvoidstorage architecture design
Core Design Contradiction:
Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The input/output chip serves as a universal interface that can accommodate multiple memory media types and host device protocols. By designing the output interface of the first chip and input interface of the second chip as standardized, the architecture can support various memory standards (SATA, SAS, NVMe, etc.) and host interfaces without requiring custom designs for each combination, thus achieving versatility while maintaining manufacturing simplicity

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

3Ease of repair

If the storage architecture uses a monolithic structure, then the system is more stable and reliable, but the ability to perform selective updates and replacements of components is limited

Engineering Contradiction:
Improvecomponent replacement capabilityVSAvoidsystem stability
Core Design Contradiction:
Ease of repairVSReliability

Solution Approach 1:

The architecture segments the storage system into independently replaceable chips connected through standardized interfaces. The first chip can be replaced to support new host device standards, the second chip can be replaced for different memory media, and the input/output chip can be updated for data transfer optimizations. This modular approach maintains system reliability through standardized connections while enabling selective component updates without replacing the entire system

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS20240273041A1Devices using chiplet based storage architectures
Publication Date: 2024.08.15 SK HYNIX INC
  • US20240273041A1 patent drawing
  • US20240273041A1 patent drawing
  • US20240273041A1 patent drawing

AI summary

A device for implementing a storage architecture includes a front-end chip configured to perform first interfacing with a first device, a plurality of back-end chips configured to perform second interfacing with second devices, and an input/output chip disposed to be separated from the front-end chip and the plurality of back-end chips and configured to perform a communication between the second devices and the plurality of back-end chips.