Virtualized Network Function Installation Profile Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for installing virtualized network functions in distributed communication architectures face scalability issues, particularly in massively distributed infrastructures, where the installation process can take hours, impacting reactivity and quality of service due to centralized resource management and lack of consideration for specific network function requirements.
Innovation Solution
A method that involves obtaining specific placement parameters for virtualized network functions, determining suitable data centers based on these parameters, and sending modified profiles for installation, allowing for the selection of data centers that match the function's needs, including compatibility and resource constraints, and incrementing counters for failed installations to manage urgent strategies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized resource management is used to install virtualized network functions, then installation decisions can be made based on global view of resources, but the installation process becomes slow and non-scalable in massively distributed infrastructures
Solution Approach 1:
The patent segments the centralized resource management into distributed components by introducing local resource managers at each data center that can autonomously make installation decisions based on local conditions and received requests, eliminating the single-point bottleneck while maintaining coordinated resource allocation across the distributed infrastructure
2Reliability
If virtualized network functions are deployed in distributed data centers on the edge of the network, then latency constraints are met and service quality is improved, but resource capacities are limited compared to centralized data centers
Solution Approach 1:
The patent enables data centers to serve multiple functions by allowing virtualized network functions to be dynamically allocated across different data centers based on service requirements. A single data center can function as both an edge location for low-latency services and be part of the broader resource pool for capacity-intensive functions, maximizing resource utilization across the distributed infrastructure
3Reliability
If a centralized controller maintains global and real-time view of available infrastructure resources, then optimal installation decisions can be made, but control streams are transmitted that may impact client data flow and quality of experience
Solution Approach 1:
The patent extracts the resource management functionality from the centralized controller and distributes it to local resource managers at each data center. These local managers maintain awareness of their own resource states and can independently handle installation requests, eliminating the need for continuous control stream transmission between centralized controller and data centers, thus removing the potential impact on client data flow
4Adaptability or versatility
If offloading mechanism is used to relocate virtualized network functions to neighboring data centers, then installation flexibility is improved, but no allowance is made for compatibility of functions with specific data centers
Solution Approach 1:
The patent implements local quality by enabling each data center to declare its specific capabilities, constraints, and compatibility characteristics. When virtualized network functions are offloaded to neighboring data centers, the selection process takes into account these local qualities, ensuring that functions are placed in data centers that are actually compatible with their requirements rather than simply selecting any available neighboring location
Data Source
AI summary
A method is described for installing a virtualized network function. The method is implemented in a service operating entity, and contributes to the implementation of a communications service, in a first data center of a group of data centers of a virtualized communications architecture, on the basis of a profile defined by a set of virtualized network function placement parameters specific to the type of virtualized network function. The profile, obtained from a profile management entity, is used to determine the first data center of the group. The identifier of the data center is then added to the profile and is transmitted to an administration entity of the virtualized architecture with a request for installation of the virtualized function in the determined first data center.