Software-Defined Networking Device for Service Profile Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional network management systems become complex with the rapid growth of mobile devices and emergence of cloud-based virtualization services, as they require manual manipulation of network topology and management of multiple devices, leading to inefficiencies in traffic handling and service optimization.
Innovation Solution
A software-defined networking device that segments physical network nodes into service profile groups based on spatiality and temporality of packets, using deep packet inspection to classify and group nodes, and configures a virtual network for efficient service-oriented network management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional network management systems are used to manage physical network nodes, then network topology can be maintained, but device complexity and management difficulty increase with rapid growth of mobile devices and cloud-based virtualization services
Solution Approach 1:
The patent segments physical network nodes into service profile groups based on learned spatial and temporal packet characteristics. This segmentation transforms the management approach from handling individual physical nodes to managing groups of nodes with similar service profiles, thereby reducing management complexity while maintaining adaptability to changing traffic patterns.
Solution Approach 2:
The patent introduces a software-defined networking device as an intermediary between the physical network infrastructure and the management system. This device learns packet characteristics and automatically segments nodes into service profile groups, acting as a mediator that simplifies management complexity while enabling adaptive network management through intelligent grouping.
2Productivity
If manual manipulation of network topology is performed, then network structure can be controlled, but traffic handling efficiency and service optimization deteriorate due to manual intervention requirements
Solution Approach 1:
The software-defined networking device performs self-service by automatically learning packet spatial and temporal characteristics and segmenting network nodes into service profile groups without manual intervention. This automation improves traffic handling efficiency while eliminating the need for manual topology manipulation, as the system adapts autonomously to traffic patterns.
Solution Approach 2:
The patent implements feedback mechanisms where the software-defined networking device continuously monitors packet characteristics and uses this feedback to dynamically segment nodes into appropriate service profile groups. This closed-loop feedback system enhances traffic handling efficiency by automatically optimizing network paths based on real-time traffic patterns without requiring manual operation.
3Adaptability or versatility
If physical network nodes are managed individually, then detailed control is achieved, but scalability deteriorates with rapid growth of mobile devices and virtualization services
Solution Approach 1:
The patent merges individual physical network nodes into service profile groups based on learned packet characteristics. By combining multiple individual nodes into unified groups with similar service profiles, the system achieves scalability to growing networks while reducing the complexity of managing each node individually, as management operations can be applied to groups rather than individual nodes.
Data Source
AI summary
Disclosed is a software-defined networking device based on service profile learning. The device includes a physical network configuration unit for configuring a physical network via identification of a plurality of physical nodes, a service profile generating unit for analyzing packets respectively communicated from end nodes among the plurality of physical nodes to generate a service profile, a service profile learning unit for learning spatiality and temporality of the packets, and segmenting the plurality of physical nodes into service profile groups based on the learning result, and a software-defined networking unit for configuring a virtual network on a service profile group basis to generate a software-defined network.