Network Tapestry Overlay for Multi-Protocol QoS Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network technologies face challenges in efficiently connecting diverse devices across various network fabrics with varying quality of service (QoS) requirements, including high bandwidth and low latency demands, while addressing limitations of IPv4 address exhaustion, network protocol inefficiencies, and high deployment costs of InfiniBand solutions.
Innovation Solution
A system for connecting devices via a virtual global network (GVN) that integrates multiple network fabrics into an end-to-end network tapestry, utilizing Advanced Smart Routing (ASR) to automatically adjust to changing conditions, offering a low-cost, plug-and-play solution with secure network optimization (SNO) and seamless integration of InfiniBand over long distances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If InfiniBand is deployed to achieve high performance connectivity, then bandwidth and latency performance is improved, but deployment cost increases
Solution Approach 1:
The network fabric is segmented into multiple protocol domains (InfiniBand, Ethernet, Fiber Channel) that can be independently deployed and managed. This allows organizations to use InfiniBand only where high performance is critical while using cheaper Ethernet elsewhere, resolving the cost-performance contradiction through spatial segmentation of the network architecture.
Solution Approach 2:
The patent creates a universal network fabric that can handle multiple protocols (InfiniBand, Ethernet, Fiber Channel) through a common overlay network. This multi-functional approach allows a single network infrastructure to provide both high-performance InfiniBand connectivity and standard Ethernet connectivity, eliminating the need for separate dedicated infrastructures and reducing overall deployment costs.
2Adaptability or versatility
If multiple network protocols are integrated into a unified fabric, then versatility and adaptability are improved, but system complexity increases
Solution Approach 1:
The patent introduces an intermediary overlay network layer that mediates between different protocol domains. This intermediary fabric with its intelligent routing protocols handles the complexity of multi-protocol integration, while presenting a simplified, unified interface to end devices. The intermediary layer translates and routes traffic between InfiniBand, Ethernet, and Fiber Channel protocols without requiring end devices to understand multiple protocols directly.
Solution Approach 2:
The network architecture uses a nested structure where protocol-specific underlay networks are embedded within a universal overlay network fabric. Each protocol domain (InfiniBand, Ethernet, Fiber Channel) operates as a nested layer with its own optimization, while the outer overlay layer provides unified routing and management. This nesting allows complex multi-protocol functionality while maintaining simplicity at each individual layer.
3Adaptability or versatility
If IPv4 addressing is used to connect devices, then compatibility with existing infrastructure is maintained, but address exhaustion limits network scalability
Solution Approach 1:
The patent transitions from the traditional flat IPv4 addressing space to a multi-dimensional addressing scheme enabled by IPv6. The 128-bit IPv6 address space provides not just more addresses but a hierarchical structure that enables new dimensions of network organization, including finer-grained subnetting, location-based addressing, and device identity integration. This dimensional expansion allows the network to scale to accommodate the Internet of Things while maintaining structured address management.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for connecting devices via a virtual global network across network fabrics using a network tapestry are disclosed. The network system comprises a first access point server in communication with a first backbone exchange server, a second access point server in communication with a second backbone exchange server, and a network tapestry comprising a first communication path connecting the first and second access point servers and a second communication path connecting the first and second backbone exchange servers.