Distributed Route Manager for Edge Router Offloading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Content providers and CDN service providers face significant strain on their computing resources due to high volumes of content requests from client devices, which is exacerbated by the need for routers with large memories and processing power to manage complex routing functions, leading to high costs and inefficiencies.
Innovation Solution
Implementing a distributed routing environment where route calculations and storage are offloaded from routers to a separate route manager, allowing commodity-based routers to be used, and leveraging excess computing capacity to pre-determine transit routes, thereby reducing processing and memory requirements for routers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If commercial level routing hardware components are used to support large FIBs and high processing volumes, then routing capability and processing power are improved, but device cost increases significantly
Solution Approach 1:
The patent extracts the routing decision-making function from the router hardware and places it in a separate route manager component. The router is reduced to a simple packet forwarding device that only needs to read routing decisions from a data structure, while the complex route calculation and FIB management are performed externally by the route manager using excess computing capacity.
Solution Approach 2:
The patent introduces an intermediary data structure that stores routing decisions externally, allowing the router to forward packets without maintaining a traditional FIB. This intermediary layer separates the router from the route manager, enabling the router to use simple commodity hardware while the route manager handles complex routing logic.
2Ease of manufacture
If commodity-based routing hardware components are used to reduce cost, then device cost is reduced, but FIB capacity and processing power are insufficient
Solution Approach 1:
The patent merges the route calculation function with the content delivery network's excess computing capacity. Instead of requiring dedicated high-power routing hardware, the system combines the router's simple forwarding function with the route manager's calculation function, utilizing otherwise wasted computing resources to achieve high-capacity routing with commodity hardware.
Solution Approach 2:
The system uses its own excess computing capacity to perform route calculations that would otherwise require specialized hardware. The route manager leverages the CDN's available processing power to calculate and maintain routing decisions, making the system self-sufficient and eliminating the need for expensive external routing infrastructure.
3Productivity
If routers maintain large FIBs to support high volume routing, then routing capacity is improved, but memory requirements and device complexity increase
Solution Approach 1:
The patent extracts the FIB data structure from the router and places it in an external route manager. The router no longer needs to maintain or search large routing tables, instead simply reading pre-calculated routing decisions from a simplified data structure. This extraction eliminates the memory burden and complexity from the router while maintaining high routing capacity through the external route manager.
Data Source
AI summary
A system, method and computer-readable medium for the determination of a transit route by a computing system distinct from a router are provided. By pre-determining routes for content within a routing environment, routers on the network forward communications according to the pre-determined route. Specifically, by distributing at least some of the route calculations functionality and/or routing entries storage/maintenance from router components (e.g., edge routers), the processing and memory requirements of such router components within the routing environment can be reduced. In one aspect, the distribution of at least a portion of the routing calculation function and routing entry storage/maintenance within a routing environment facilitates the use of lower-capability, cheaper commodity-based routers.