VNF Provisioning via Performance-Based Tuple Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In service provider environments, scaling network services on purpose-built hardware is expensive and time-consuming, limiting operational and capital efficiency, while existing virtualization solutions lack a systematic approach for provisioning network services based on performance characteristics of Virtual Network Functions (VNFs) and Virtual Machines (VMs.
Innovation Solution
A method and system that select and provision network services by determining suitable VNFs and VMs, generating tuples for VNF-VM combinations, estimating performance parameters, processing scores to generate VNF Forwarding Graphs (FGs), and provisioning services based on grade scores, incorporating both VNF and VNF FG edge performance evaluations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network services are deployed on purpose-built hardware, then service reliability is improved, but scaling cost and time increase significantly
Solution Approach 1:
The patent creates virtual copies of network functions through VNFs that can be deployed on standard hardware. Instead of copying physical hardware, the system instantiates virtualized network function components that replicate the functionality of traditional network devices, enabling rapid provisioning and scaling while maintaining service reliability through software-defined abstractions.
Solution Approach 2:
The patent replaces mechanical/physical network function deployment with software-based virtualization. By substituting physical network function appliances with virtualized network functions running on standard servers, the system eliminates the need for specialized hardware while maintaining functional equivalence, thereby enabling faster scaling and reduced costs.
2Adaptability or versatility
If multiple VNFs are deployed on multiple VMs, then service flexibility and scalability are improved, but performance management complexity increases
Solution Approach 1:
The patent implements a feedback mechanism that continuously monitors VNF performance parameters and uses this information to dynamically select and provision optimal VNF-VM combinations. The system collects performance data from deployed VNFs, analyzes it against service requirements, and adjusts provisioning decisions accordingly, creating a closed-loop system that manages complexity through intelligent feedback-driven automation.
Solution Approach 2:
The patent manages complexity by dynamically changing performance parameters such as CPU allocation, memory allocation, and network bandwidth for VNFs based on actual service demands. The system adjusts these parameters in response to performance measurements and service level requirements, enabling flexible resource management without manual intervention for each parameter change.
3Measurement precision
If VNF performance parameters are estimated for each tuple, then provisioning accuracy is improved, but computational processing time increases
Solution Approach 1:
The patent performs preliminary estimation of VNF performance parameters for all possible VNF-VM tuples before actual service provisioning. By pre-calculating and storing performance estimates for each tuple in advance, the system avoids time-consuming real-time calculations during provisioning events, thus maintaining high accuracy while reducing actual provisioning time through preparatory computations.
Data Source
AI summary
A method includes performing, by a processor: receiving a selection of a network service, determining a plurality of Virtual Network Functions (VNFs) and a plurality of Virtual Machines (VMs) based on the selection of the network service, the plurality of VNFs being configured to run on the plurality of VMs, generating a plurality of tuples, each of the plurality of tuples corresponding to a combination of one of the plurality of VNFs and one of the plurality of VMs where the one of the plurality of VNFs is configured to run on the one of the plurality of VMs, receiving a plurality of VNF performance parameters, estimating values for each of the plurality of VNF performance parameters for each of the plurality of tuples, respectively, processing the values for each of the plurality of VNF performance parameters to generate a plurality of scores for the plurality of tuples, respectively, generating a plurality of VNF Forwarding Graphs (FGs), each of the VNF FGs corresponding to a combination of ones of the plurality of tuples that support the network service, generating a plurality of grade scores for the plurality of VNF FGs, respectively, based on the plurality of scores for the plurality of tuples, and provisioning the network service on one of the plurality of VNF FGs based on the plurality of grade scores.