Distributed Memory Routing via Network Switch Mapping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing distributed storage systems face challenges in achieving high performance and low latency while managing complexity, scalability, and cost, due to reliance on centralized architectures and inefficient use of network resources.

Innovation Solution

A distributed memory system that integrates end-host and network resources using a network component with a switching device and mapping resource, allowing direct routing of memory requests based on data addresses with embedded identification information, and leveraging existing network protocols to manage storage and redundancy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a centralized storage architecture is used, then ease of management and simplicity are improved, but performance and scalability deteriorate due to a single device bottleneck

Engineering Contradiction:

Improveease of managementVSAvoidperformance

Core Design Contradiction:

Ease of operationVSProductivity

Solution Approach 1:

The patent segments the centralized storage architecture into multiple distributed storage nodes that independently handle storage operations. Each node maintains local state and can service requests autonomously, eliminating the single-device bottleneck while preserving management simplicity through standardized interfaces.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimensional approach by embedding routing logic directly within network switches rather than relying on end-host software. This shifts the architecture from a software-centric management model to a network-centric model, improving performance by reducing software overhead while maintaining ease of management through hardware-enforced routing rules.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Productivity

If distributed storage is implemented, then performance and cost are improved, but system complexity increases due to distributed state management

Engineering Contradiction:

ImproveperformanceVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent extracts the complex distributed state management logic from end-host software and relocates it to network switch hardware. This extraction eliminates the need for complex software-based state tracking at hosts while maintaining the performance benefits of distributed architecture through hardware-enforced routing and state consistency.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces network switches as intermediaries between end hosts and storage nodes. These switches embed routing logic that automatically manages distributed state, acting as a mediator that simplifies the interaction between hosts and storage while maintaining system-wide consistency without requiring complex host-side software.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If end-host software is used for storage management, then flexibility and adaptability are improved, but latency and resource utilization deteriorate

Engineering Contradiction:

ImproveflexibilityVSAvoidlatency

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent substitutes software-based storage management with hardware-based network switch logic. This replacement eliminates software processing overhead and context switching that cause latency, while maintaining flexibility through programmable switch architectures that can be configured for different storage protocols and access patterns.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent enables network switches to autonomously make routing decisions based on embedded routing logic, eliminating the need for continuous software intervention at end hosts. This self-service approach reduces latency by handling storage management operations directly in network hardware while maintaining adaptability through configurable routing rules.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11271893B1Systems, methods and devices for integrating end-host and network resources in distributed memory

Publication Date: 2022.03.08 OPEN INVENTION NEWTORK LLC

AI summary

Systems, methods and devices for distributed memory management comprising a network component configured for network communication with one or more memory resources that store data and one or more consumer devices that use data, the network component comprising a switching device in operative communication with a mapping resource, wherein the mapping resource is configured to associate mappings between data addresses associated with memory requests from a consumer device relating to a data object and information relating to a storage location in the one or more memory resources associated with the data from the data object, wherein each data address has contained therein identification information for identifying the data from the data object associated with that data address; and the switching device is configured to route memory requests based on the mappings.