Centralized Delay-Based Traffic Rate Control in Split-Architecture Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional network architectures with tightly coupled forwarding and control planes face complexity and slow advancements in control plane mechanisms, limiting effective congestion control in split-architecture networks, which are often managed through distributed processes that result in non-optimal bandwidth sharing and inefficient congestion management.
Innovation Solution
Implementing a centralized traffic rate control process in split-architecture networks using a central controller to monitor congestion across the network, calculate throttling rates based on path delays, and apply these rates to ingress switches to manage congestion efficiently and fairly, prioritizing traffic groups based on assigned weights.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a split-architecture network is implemented to simplify switches and shift intelligence to controllers, then the forwarding plane performance is improved, but the control plane complexity increases
Solution Approach 1:
The network architecture is segmented into separate forwarding plane (switches) and control plane (controllers), allowing independent optimization of each. The forwarding plane handles high-speed packet forwarding while controllers manage control logic, enabling performance improvement in forwarding without being constrained by control plane capabilities.
Solution Approach 2:
A standardized communication protocol (OpenFlow) acts as an intermediary between switches and controllers, providing a clean interface that simplifies the control plane while maintaining flexibility. This protocol layer abstracts the complexity of controller-switch communication and enables modular design.
2Adaptability or versatility
If distributed congestion control processes are used in split-architecture networks, then autonomy of network elements is improved, but congestion management efficiency deteriorates
Solution Approach 1:
The system implements feedback mechanisms where controllers monitor network congestion status and adjust throttling rates dynamically based on real-time conditions. This allows distributed elements to operate autonomously while maintaining efficient congestion control through continuous feedback loops that adapt to changing network states.
Solution Approach 2:
Throttling rates are made dynamic rather than static, allowing the system to adapt congestion control parameters in real-time based on network conditions. This enables efficient congestion management while preserving the autonomy of distributed network elements to respond locally to changes.
3Manufacturing precision
If delay-based throttling rates are calculated and applied at ingress switches, then congestion control precision is improved, but network-wide coordination complexity increases
Solution Approach 1:
Throttling rates are calculated and applied at the ingress switches before traffic enters the network core, enabling early congestion control. This preliminary action prevents congestion from developing in the first place, improving precision while simplifying coordination by handling control decisions at the network edge rather than requiring complex end-to-end coordination.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A process is performed by a controller in a split-architecture network. The controller monitors congestion of traffic groups across the split-architecture network and executes the process to provide delay based data rate control to alleviate congestion of the traffic groups. The process includes configuring an ingress switch and egress switch for each traffic group to collect delay measurement data for data packets of each traffic group as they arrive at the ingress switch and egress switch. The delay measurement data is received from the ingress switch and egress switch of each traffic group. A check is made whether a minimum data packet delay for any traffic group exceeds a defined threshold value. A throttling rate is calculated for each traffic group in the split-architecture network, in response to the defined threshold value being exceeded.