SDN Controller Scheduling Non-Real-Time Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional network architectures face congestion during peak hours due to inadequate resource management, leading to decreased user experience and increased construction costs, as software-defined networks (SDNs) can only balance resource utilization within existing network limits, failing to address peak-hour traffic surges effectively.
Innovation Solution
A scheduling method and system that pre-schedules non-real-time traffic transmission by determining optimal forwarding paths and sending times based on statistical network traffic information, reducing network pressure and congestion by selecting paths with maximum link bandwidth and lowest bandwidth utilization rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network resources are improved to handle peak-hour traffic, then network congestion during peak hours is reduced, but network construction costs increase greatly
Solution Approach 1:
The patent applies preliminary action by pre-scheduling non-real-time traffic during off-peak hours before peak hours occur. The scheduling system analyzes historical traffic patterns and proactively transfers non-real-time data (such as file transfers, backups, and software updates) to different storage locations or content sources during low-utilization periods. This preparatory action ensures that when peak hours arrive, the network infrastructure doesn't need to handle excessive traffic loads, thereby maintaining service quality without requiring costly infrastructure expansions.
2Reliability
If network resources are improved to handle peak-hour traffic, then network congestion during peak hours is reduced, but network resources are wasted most of the time
Solution Approach 1:
The patent implements dynamics by creating a flexible, adaptive scheduling system that dynamically adjusts traffic distribution based on real-time network conditions and historical patterns. The system continuously monitors network utilization, identifies non-real-time traffic flows, and dynamically schedules their transmission during periods of low network utilization. This dynamic approach ensures that network resources are optimally utilized across different time periods, preventing waste during off-peak hours while ensuring adequate capacity during peak hours, thereby eliminating the need for permanent over-provisioning of network resources.
3Productivity
If software defined network schedules network resources, then resource utilization is balanced within network range, but network congestion still exists when peak-hour traffic exceeds total network resources
Solution Approach 1:
The patent applies preliminary action by pre-scheduling non-real-time traffic during off-peak hours before peak hours occur. The scheduling system analyzes historical traffic patterns and proactively transfers non-real-time data (such as file transfers, backups, and software updates) to different storage locations or content sources during low-utilization periods. This preparatory action ensures that when peak hours arrive, the network infrastructure doesn't need to handle excessive traffic loads, thereby maintaining service quality without requiring costly infrastructure expansions.
4Productivity
If non-real-time traffic is pre-scheduled based on statistical information, then traffic peaks are reduced and network pressure decreases, but real-time traffic requirements may be affected
Solution Approach 1:
The patent applies local quality by differentiating between various types of traffic and applying different scheduling strategies to each category. The system identifies and separates non-real-time traffic (such as file transfers, backups, and software updates) from real-time traffic (such as voice over IP, video conferencing, and online gaming). Priority markers and traffic classification mechanisms ensure that real-time traffic receives immediate processing and is not delayed by pre-scheduled non-real-time operations. This localized approach to traffic management allows aggressive pre-scheduling of non-critical traffic while preserving the quality and timeliness of real-time services.
Data Source
AI summary
The application provides a scheduling method, apparatus and system. A controller receives scheduling request information sent by a content source. The scheduling request information includes a constraint condition of non-real-time traffic transmission, and the constraint condition includes: information about a first address and information about a second address. A first forwarding path and a sending time for sending the traffic are determined according to the information about the first address, the information about the second address, and statistical information about network traffic. The controller sends a first message that includes a sending rate to the content source at the sending time, and the traffic is sent through the first forwarding path.