Carrier-Grade Router Route Segmentation for Scalable FIBs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Carrier-grade routers face scalability issues due to full routing tables consuming critical resources and limiting further growth, as existing network operating system designs program every route into each linecard, leading to resource constraints and inefficiencies.
Innovation Solution
The routing table is split based on utilization within a distributed, disaggregated cluster (DDC) router, with high-volume routes stored on one set of linecards and low-volume routes on another, allowing for efficient resource allocation and scalability without altering the network control plane or forwarding behaviors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If every route is programmed into each linecard, then routing completeness is maintained, but device resources are consumed and scalability is limited
Solution Approach 1:
The routing table is segmented into two parts: a first portion stored in the forwarding information base (FIB) on the linecard, and a second portion stored in a separate data structure in memory. This segmentation allows frequently accessed routes to remain in fast FIB memory while less frequently accessed routes are stored elsewhere, reducing linecard resource consumption while maintaining complete routing capability.
2Quantity of substance
If routing table size is increased to handle more routes, then route handling capacity improves, but resource constraints are exceeded
Solution Approach 1:
Different portions of the routing table are stored in different locations with different access characteristics. The first portion in FIB provides fast access for common routes, while the second portion in memory accommodates additional routes. This local quality differentiation enables the system to handle more total routes without uniformly increasing resource consumption across all linecards.
3Adaptability or versatility
If routing information is distributed across multiple linecards, then scalability improves, but system complexity increases
Solution Approach 1:
A route reflection server acts as an intermediary between the control plane and the data plane. It receives routing information, determines which routes belong in the first portion versus the second portion, and distributes them appropriately. This intermediary simplifies the overall system architecture by centralizing the decision-making logic for routing information distribution, rather than requiring complex peer-to-peer coordination between multiple linecards.
Data Source
AI summary
Aspects of the subject disclosure may include, for example, a method of receiving, by a processing system including a processor, a plurality of routes in a network through a routing device having a fabric, a first set of distributed chassis packet forwarders (DCPs) that support external service interfaces and a second set of DCPs; storing, by the processing system, the plurality of routes in a forwarding information base (FIB); determining, by the processing system, a third set of routes in the FIB having a high volume of usage; determining, by the processing system, a fourth set of routes in the FIB having a low volume of usage; copying, by the processing system, the third set of routes to the first set of DCPs; and copying, by the processing system, the fourth set of routes to the second set of DCPs. Other embodiments are disclosed.


