Super Controller Latency Distribution for SDN
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Software Defined Network (SDN) cross-domain services do not support deployment of services with end-to-end latency restrictions, as existing methods for latency distribution across autonomous systems are static and inappropriate, failing to adjust based on network and service status.
Innovation Solution
A method for establishing a forwarding path in a network using a super controller (SC) that manages multiple domain controllers (DCs), where the SC receives a service request, selects a latency distribution manner based on the service type, divides the latency into periods corresponding to each DC, and triggers each DC to establish a forwarding path segment, connecting them to form a customized latency path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a static latency distribution method is used across autonomous systems, then the network structure is simple to manage, but the service cannot meet dynamic latency requirements of different services
Solution Approach 1:
The patent segments the end-to-end latency requirement into multiple domain-specific latency requirements. The super controller divides the total latency budget into several parts, assigning each part to a specific domain controller based on service types and network conditions. This allows each domain to independently optimize its latency distribution while collectively meeting the overall service requirement.
Solution Approach 2:
The patent implements dynamic latency distribution where the super controller continuously monitors network status and service requirements, adjusting latency allocations in real-time. Different service types (e.g., real-time video, file transfer) receive different latency distributions based on current network conditions, enabling the system to adapt to changing requirements without manual reconfiguration.
2Productivity
If a uniform latency distribution is applied to all domains, then the implementation is simple, but network utilization is not optimized
Solution Approach 1:
The patent applies local quality by allowing each domain to have customized latency distribution characteristics based on its specific network conditions and service requirements. The super controller assigns different latency weights to different domains according to their capabilities and current status, enabling optimal local utilization while maintaining global coordination.
Solution Approach 2:
The system dynamically changes latency distribution parameters based on service types and network conditions. The super controller adjusts latency allocation parameters in real-time, transforming the static uniform distribution into a dynamic adaptive distribution that optimizes network utilization for different scenarios.
3Extent of automation
If manual latency distribution configuration is used, then the system is easy to understand, but it cannot respond to changing network status and service requirements
Solution Approach 1:
The patent implements a feedback mechanism where the super controller continuously monitors network status, service performance, and domain conditions. Based on this feedback, the controller automatically adjusts latency distributions and reconfigures service paths to meet changing requirements, enabling the system to self-optimize without manual intervention.
Solution Approach 2:
The system enables automatic self-service through the super controller that autonomously performs latency distribution optimization. The controller automatically identifies service types, monitors network status, calculates optimal latency distributions, and configures domain controllers accordingly, reducing the need for manual operation while improving adaptability.
Data Source
AI summary
The application discloses a method for establishing a forwarding path in a network, where the network includes a SC and a plurality of DCs, the SC manages the plurality of DCs, each of the plurality of DCs manages one domain, and the method includes: receiving, by the SC, a service request message; selecting, by the SC, a latency distribution manner based on the service request message; dividing, by the SC, a first latency into a plurality of latency periods; sending, by the SC, each of the plurality of latency periods to a DC corresponding to the latency period, and triggering each of the plurality of DCs to establish a forwarding path segment in the domain managed by the DC; and connecting, by the SC, the plurality of forwarding path segments established by the plurality of DCs, to form a forwarding path.