Flow-Channel Routing for Link Failure and Congestion Response
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network architectures face challenges in scalability, versatility, and efficiency due to increasing network load and diverse traffic types, particularly in high-performance computing and media streaming, with conventional congestion control mechanisms being slow and ineffective.
Innovation Solution
A network architecture utilizing flow channels with dynamic, distributed flow management and adaptive routing, implemented through a custom ASIC switch chip, enables 'on the fly' rerouting and congestion control, ensuring efficient traffic handling and minimizing buffer overflow.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional network architectures are used to handle increasing network loads and diverse traffic types, then network scalability and versatility are compromised, but implementing complex routing mechanisms increases device complexity
Solution Approach 1:
The patent implements dynamic routing decisions where switches can change their forwarding paths in real-time based on detected errors or congestion conditions. The system transitions from static routing tables to dynamic path selection, allowing networks to adapt to changing conditions without requiring complex pre-configured routing mechanisms for every possible scenario.
Solution Approach 2:
The routing mechanism operates autonomously by detecting errors and congestion conditions locally at each switch, then automatically adjusting paths without requiring centralized control or complex external management. The system self-regulates based on real-time network conditions, reducing the need for complex external routing management.
2Reliability
If flow channels are used for congestion control and on the fly routing, then network utilization and reliability are improved, but the system complexity increases
Solution Approach 1:
The patent segments network traffic into distinct flow channels, allowing different types of traffic to be managed independently. Each flow channel can be controlled separately for congestion management and error handling, which simplifies the overall system by breaking down complex routing decisions into manageable per-flow operations rather than requiring complex global routing algorithms.
Solution Approach 2:
The system implements feedback mechanisms where switches monitor network conditions (errors, congestion) and adjust routing decisions based on this feedback. This closed-loop approach simplifies complexity by using local feedback to drive automatic adjustments, eliminating the need for complex predictive algorithms or centralized control systems.
3Productivity
If adaptive routing is implemented to avoid failing links, then network efficiency under heavy loads is improved, but packet delivery order may be compromised
Solution Approach 1:
The patent segments packets into flow channels based on their source-destination pairs and prioritization requirements. This segmentation ensures that packets within the same flow channel maintain their relative order while different flow channels can be routed independently for optimization. The segmentation at the flow level rather than packet level preserves delivery order while enabling adaptive routing.
Solution Approach 2:
The system applies different routing strategies to different flow channels based on their specific requirements. High-priority flows maintain strict ordering and use deterministic routing, while lower-priority flows can use more flexible adaptive routing. This local differentiation allows the system to optimize overall efficiency without compromising packet order for critical flows.
Data Source
AI summary
Systems and methods are provided for “on the fly” routing of data transmissions in the presence of errors. Switches can establish flow channels corresponding to flows in the network. In response to encountering a critical error on a network link along a transmission path, a switch can generate an error acknowledgement. The switch can transmit the error acknowledgements to ingress ports upstream from the network link via the plurality of flow channels. By transmitting the error acknowledgement, it indicates that the network link where the critical error was encountered is a failed link to ingress ports upstream from the failed link. Subsequently, each ingress port upstream from the failed link can dynamically update the path of the plurality of flows that are upstream from the failed link such that the plurality of flows that are upstream from the failed link are routed in a manner that avoids the failed link.