Multi-Layer Virtual Topology for SDN Energy Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data center networks with 'rich-connection' architectures, such as Fat-Tree, face low energy resource utilization at low loads due to inefficient energy management, leading to unnecessary energy consumption and increased complexity in existing energy-aware routing methods.
Innovation Solution
An energy-efficient traffic scheduling algorithm based on multi-layer virtual topologies (EMV-SDN) is introduced, which uses integer linear programming to minimize network energy consumption by dynamically adjusting the state of switches and links based on current load conditions, optimizing the routing of data flows through a multi-layer virtual topology structure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If ECMP routing algorithm is used to schedule data flows in Fat-Tree data center network, then load balancing is achieved, but energy consumption is not optimized and network devices cannot be made dormant
Solution Approach 1:
The patent transforms the static ECMP routing algorithm into a dynamic energy-aware routing algorithm. The SDN controller dynamically adjusts routing decisions based on real-time network load conditions, switching between active and dormant states for network devices. This dynamic adaptation allows the system to optimize energy consumption while maintaining load balancing capabilities.
Solution Approach 2:
The patent changes the routing algorithm by introducing energy consumption as a new parameter alongside traditional metrics like path length and load balance. The modified routing algorithm considers multiple parameters including energy efficiency, path cost, and current network state, enabling flexible routing decisions that optimize both performance and energy consumption.
2Reliability
If all network switches are kept running at 100% power to ensure reliability and availability, then network reliability is maintained, but energy consumption increases unnecessarily
Solution Approach 1:
The patent implements dynamic power state management for network switches, allowing them to transition between active and dormant states based on real-time load conditions. The SDN controller monitors network traffic and dynamically adjusts switch states, ensuring reliability during high load while reducing energy consumption during low load periods.
Solution Approach 2:
The network system automatically adjusts its own power consumption based on demand. The SDN controller enables network devices to self-manage their operational states, switching between active and dormant modes according to actual network requirements, thereby eliminating the need for continuous full-power operation.
3Use of energy by moving object
If multi-layer virtual topology is introduced to optimize energy efficiency, then energy consumption is reduced, but system complexity increases
Solution Approach 1:
The patent introduces an SDN controller as an intermediary that manages the complexity of multi-layer virtual topology. The controller handles the intricate tasks of monitoring network state, calculating optimal paths, and managing virtual topology layers, thereby reducing the complexity burden on individual network devices while enabling energy optimization.
Solution Approach 2:
The patent creates virtual copies of the network topology at different abstraction layers. These virtual topologies represent simplified views of the physical network, allowing the SDN controller to perform energy optimization calculations on virtual models without directly manipulating the complex physical infrastructure.
4Use of energy by moving object
If centralized SDN control is implemented to enable energy-efficient scheduling, then energy optimization is achieved, but control plane complexity increases
Solution Approach 1:
The patent segments the control functions into modular components within the SDN controller. The controller is divided into distinct modules for monitoring, path calculation, routing decision-making, and state management. This segmentation reduces individual module complexity while enabling comprehensive energy optimization through coordinated operation of all modules.
Data Source
AI summary
An energy-efficient traffic scheduling algorithm based on multiple layers of virtual sub-topologies is provided. First, a mathematical optimization model for an energy-efficient traffic scheduling problem is established, to minimize network energy consumption while ensuring the capability of bearing all network data flows. Then, the mathematical optimization model is resolved using an energy-efficient traffic scheduling algorithm based on a multi-layer virtual topology, to obtain an energy-efficient scheduling scheme of the data flows. The virtual topology and switch ports in an upper layer are made dormant to save energy. The method can dynamically adjust the working state of the virtual sub-topology in the upper layer according to current link utilization. A path with a minimum number of hops and lowest maximum link utilization can be found in the booted sub-topology, to route the data flow, solving the problem that a “rich-connection” data center network has low energy resource utilization at low load.