Network Scheduling via SDN Traffic Profiling for Energy-Latency Trade-offs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current energy conservation methods in communications networks are inefficient in managing varying network traffic parameters such as power consumption, latency, bandwidth, and security, leading to suboptimal energy savings and network performance.
Innovation Solution
A traffic scheduling system that utilizes software-defined networking (SDN) to profile and analyze network traffic, generating scheduling plans that configure network resources to meet specific communication needs by optimizing idle periods and energy states of components like physical layer, MAC layer, and forwarding resources, thereby enhancing energy efficiency and reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by stationary object
If network resources are placed in low-power state to save energy, then energy consumption is reduced, but network traffic latency increases
Solution Approach 1:
The patent implements dynamic power state transitions for network resources based on real-time traffic conditions. The system monitors network traffic parameters and dynamically adjusts power states of network resources (active, idle, low-power) to match actual demand, resolving the contradiction between energy savings and latency by making power state selection adaptive rather than static
Solution Approach 2:
The system changes power state parameters (P_state) of network resources based on traffic intensity and latency requirements. By adjusting the power state parameter dynamically according to traffic conditions, the system achieves energy savings during low-traffic periods while maintaining low latency during high-traffic periods, thus resolving the contradiction
2Use of energy by stationary object
If wake time is extended to increase energy savings, then energy consumption is reduced, but network traffic latency increases
Solution Approach 1:
The patent makes wake time dynamic rather than fixed, adjusting it based on monitored network traffic patterns and latency requirements. The system learns from traffic history and adapts wake times to balance energy savings with acceptable latency, resolving the contradiction by making the power management policy adaptive to actual network conditions
Solution Approach 2:
The system implements feedback mechanisms that monitor network traffic and latency performance, using this information to adjust wake times and power state transitions. This closed-loop control allows the system to optimize the trade-off between energy savings and latency by continuously adapting to actual network conditions
3Adaptability or versatility
If network resources are configured to handle multiple network parameters (bandwidth, availability, security), then communication needs are met, but device complexity increases
Solution Approach 1:
The patent implements a universal power management framework that handles multiple network parameters (bandwidth, availability, security) through a single integrated system. The framework provides multi-functional capabilities to manage diverse communication requirements using unified policies and procedures, reducing the effective complexity despite handling multiple parameters
Solution Approach 2:
The system segments the management of different network parameters into separate modular components within the power management framework. Each parameter (bandwidth, availability, security) can be managed independently through dedicated modules, allowing complex multi-parameter configuration to be broken down into manageable segments that can be configured and optimized separately
Data Source
AI summary
In one implementation, a network scheduling method includes collecting information describing pending traffic for a network and generating a plurality of initial scheduling plans for network resources associated with the pending traffic within the network. The network scheduling method also includes analyzing a plurality of scheduling plans derived from the plurality of initial scheduling plans, and defining a current scheduling plan for the network resources from the plurality of scheduling plans.