Dynamic Network Node Function Placement for Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network management solutions cannot optimize the placement of server functions to improve client-server communication quality and resource usage, leading to suboptimal network performance and resource utilization, especially in LTE networks and cloud computing environments.
Innovation Solution
A method is proposed to enable network node functions on a second network node by determining a resource provider entity based on client identification data, obtaining client context data, and transmitting a function placement request to the second network node, allowing for the redirection of client requests for improved latency and resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network traffic is managed using traditional network management solutions (MPLS, VPNs, DiffServ, IntServ), then network traffic can be prioritized and paths can be assigned to specific endpoints, but the location of client and server endpoints cannot be changed, preventing optimization of network resources and reduction of network overhead
Solution Approach 1:
The patent implements dynamic service function placement where network services can be moved between different network nodes based on real-time network conditions, client locations, and resource availability. This dynamic approach allows the system to adapt to changing conditions without requiring complex static configuration management, resolving the contradiction between maintaining service quality and reducing management complexity
Solution Approach 2:
The patent introduces a service placement manager as an intermediary component that coordinates between clients, network nodes, and service functions. This intermediary handles the complexity of service placement and endpoint management, allowing traditional network management solutions to continue functioning while enabling optimized service delivery without increasing overall system complexity
2Productivity
If cloud service providers do not allow network service providers to optimize cloud service delivery, then cloud services can be delivered without network-specific optimization, but optimal service delivery considering network conditions cannot be achieved
Solution Approach 1:
The patent segments the service delivery system into independent components: cloud service providers, network service providers, and service placement managers. Each entity maintains its independence and control over its resources while collaborating through standardized interfaces. This segmentation allows network service providers to optimize service delivery without requiring cloud service providers to expose their internal infrastructure, maintaining both productivity and adaptability
Solution Approach 2:
The patent creates a universal service placement mechanism that can work across different cloud service provider environments without requiring proprietary integrations. The service placement manager uses standardized network information and service descriptors that can be applied universally, enabling network optimization while maintaining flexibility and compatibility with various cloud providers
3Quantity of substance
If network node functions are placed on distant network nodes, then network resources can be distributed, but network latency increases and quality of experience deteriorates
Solution Approach 1:
The patent implements local quality optimization by placing service functions on network nodes that are geographically and topologically close to the clients accessing those services. The service placement manager evaluates network topology, client locations, and resource availability to determine optimal placement locations, ensuring that services are distributed across the network while maintaining low latency by prioritizing local placement whenever possible
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention enables placement and use of a network node function in a second network node instead of using the network node function in a first network node. The network node function is e.g. a server function or a router function. The second network node is typically located in or close to the client device or to the network of an operator servicing a client device to enable optimization of client- server communication and optimization of the load on the client- server communication in the network. Client requests transmitted to the first network node may be redirected to the second network node.