Switch Forwarding Database Synchronization via Dedicated Sync Queues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches to synchronizing forwarding databases across multiple interconnected layer-2 switches in Ethernet systems face challenges such as inefficient software implementation, scalability issues, and interference with data path bandwidth, leading to packet drops and failure in providing a singular switch view for network management.
Innovation Solution
An enhanced data path forwarding approach that uses a separate queue for sync packets, transmitted over inter-switch connectivity with an invalid destination port, allowing all switches to learn MAC addresses without impacting data path traffic and requiring only one-time provisioning, thus eliminating software complexity and real-time bandwidth consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional periodic synchronization messaging is used between line cards, then forwarding database synchronization is achieved, but hardware bandwidth is excessively consumed and packet drops occur
Solution Approach 1:
The patent implements event-triggered synchronization instead of periodic synchronization. MAC address updates are synchronized only when changes occur, using event-driven messaging rather than continuous periodic updates. This eliminates unnecessary bandwidth consumption while maintaining synchronization reliability.
Solution Approach 2:
Each line card independently generates and transmits synchronization messages when it detects a MAC address change. The system uses self-service synchronization where line cards autonomously update peers without requiring centralized coordination or periodic polling, reducing overall bandwidth consumption.
2Reliability
If software-based interprocess communication is used for MAC address synchronization, then synchronization is achieved, but system complexity and processing overhead increase
Solution Approach 1:
The patent replaces software-based interprocess communication with hardware-level direct memory access (DMA) and memory-mapped I/O mechanisms. MAC address synchronization is achieved through direct hardware registers and memory structures, eliminating the need for complex software polling and message passing protocols.
3Reliability
If centralized repository approach is used for forwarding database synchronization, then synchronization is achieved, but scalability is limited and single point of failure risk increases
Solution Approach 1:
The patent segments the centralized synchronization function into distributed peer-to-peer synchronization between line cards. Each line card maintains its own forwarding database and independently synchronizes with peers, eliminating the single point of failure and enabling linear scalability as line cards are added to the system.
Solution Approach 2:
The synchronization mechanism is designed to be universally applicable across different line card types and configurations. The event-driven protocol and hardware-based implementation provide a scalable framework that works regardless of the number of line cards or their specific functions, enhancing system adaptability.
4Reliability
If periodic synchronization is implemented in multi-chassis architectures, then forwarding database consistency is maintained, but hardware bandwidth capacity is exceeded and packet drops increase
Solution Approach 1:
The patent replaces periodic synchronization with event-triggered synchronization. MAC address updates are transmitted immediately when changes occur, rather than waiting for periodic update cycles. This eliminates synchronization-induced bandwidth saturation and prevents packet drops while maintaining forwarding database consistency across chassis.
Data Source
AI summary
A method and a multi-switch architecture include learning a media access control (MAC) address at a first switch in a multi-switch architecture; storing the MAC address in a forwarding database of the first switch; transmitting a data frame to one or more switches in the multi-switch architecture via inter-switch connectivity between the first switch and the one or more switches, wherein the data frame is created to enable the one or more switches to learn the MAC address therefrom; learning the MAC address from the data frame at the one or more switches; and storing the MAC address in a forwarding database for each of the one or more switches. This further includes transmitting the data frame via ports and queues in the inter-switch connectivity that are separate from ports and queues in a data path between the first switch and the one or more switches to avoid data path interference.


