Location-Based Network Routing Reduces CAM Dependency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional computer networks face inefficiencies due to excessive broadcast traffic, redundant routing, and the need for expensive content addressable memory (CAM) in network devices, particularly in data centers where redundancy is planned to increase bandwidth and throughput.
Innovation Solution
A location-based addressing and routing scheme that embeds routing information within packet addresses, using a controller to assign address values including site prefix, controller ID, switch ID, port ID, and MAC address, allowing switches to directly route packets without additional routing information and reducing the reliance on CAM.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional routing protocols are used in data center networks, then routing flexibility and adaptability are improved, but device complexity and cost increase due to the need for expensive content addressable memory (CAM)
Solution Approach 1:
The patent segments the routing information into two parts: location information embedded in packet addresses (which contains switch ID and port ID) and routing logic implemented in simple switch fabric. This segmentation eliminates the need for complex CAM-based routing tables while maintaining routing flexibility, as switches only need to extract and act on location information from the address field.
Solution Approach 2:
The patent introduces location information as an intermediary element that mediates between the source and destination devices. This location information, embedded in the packet address, serves as a simplified routing directive that replaces complex routing table lookups, thereby reducing device complexity while preserving routing adaptability.
2Ease of operation
If broadcast traffic is used for network communication, then ease of operation and device compatibility are improved, but network efficiency deteriorates due to excessive broadcast messages and redundant traffic
Solution Approach 1:
The patent extracts the destination location information directly from the packet address field, removing the need for broadcast-based destination discovery. By embedding switch ID and port ID in the address, the system eliminates redundant broadcast traffic while maintaining ease of operation, as devices simply follow the location information already present in the packet.
3Measurement precision
If content addressable memory (CAM) is used for routing decisions, then routing precision and speed are improved, but device cost and complexity increase
Solution Approach 1:
The patent replaces expensive, complex CAM memory with simple, inexpensive switch fabric logic that extracts routing information from packet addresses. This substitution uses cheap, simple components to achieve the same routing precision, thereby reducing device complexity and cost while maintaining routing accuracy.
4Productivity
If location-based routing is implemented, then network efficiency is improved by reducing broadcast traffic, but address structure complexity increases
Solution Approach 1:
The patent merges routing information (switch ID and port ID) directly into the packet address field, combining what were previously separate functions (addressing and routing) into a unified structure. This merging eliminates the need for separate routing tables and broadcast mechanisms, improving network efficiency while the integrated nature of the solution actually simplifies overall system complexity.
Data Source
AI summary
Apparatus and method for routing packets in a computer network. A network switch for routing packets in a computer network includes a plurality of ports for communicative connection of computing devices to the switch, and routing logic. The routing logic is configured to extract, from a packet received via a first of the ports, a destination address that identifies a destination device to which the packet is directed; to extract from the destination address a switch ID value and a port ID value; to compare the switch ID value extracted from the destination address to a switch ID value identifying the network switch; and to transmit the packet via a second of the ports of the network switch corresponding to the port ID value based on the switch ID value extracted from the destination address being equal to the switch ID value identifying the network switch.