Virtual Network Resource Allocation Predictor

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In virtualized network environments, non-network related tasks often consume extensive computing resources, leading to potential degradation of network functions, as they share the same hardware resources, necessitating a solution to ensure consistent performance of critical virtualized network functions.

Innovation Solution

A method and system for determining optimized resource utilization in virtual networks by collecting computing resource parameters and key performance indicators, using a predictor engine to estimate future resource requirements, and adjusting resource allocation to prevent bottlenecks, thereby maintaining network performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If computing resources are shared between network functions and non-network tasks, then resource utilization efficiency is improved, but network function performance reliability deteriorates

Engineering Contradiction:

Improveresource utilization efficiencyVSAvoidnetwork function performance

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system dynamically adjusts resource allocation between network functions and non-network tasks based on real-time monitoring of performance metrics and predictive analytics. The resource allocation is not static but adapts continuously to changing conditions, ensuring network functions receive adequate resources while maximizing overall utilization efficiency.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary actions by predicting future resource requirements and potential performance degradation before they occur. Using machine learning models and historical data, it proactively identifies when non-network tasks may interfere with network functions and pre-emptively adjusts resource allocation to prevent performance issues.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If resource allocation is increased for network functions, then network performance reliability is improved, but resource utilization efficiency worsens

Engineering Contradiction:

Improvenetwork function performanceVSAvoidresource utilization efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system changes allocation parameters dynamically based on actual network conditions and demand. Rather than maintaining fixed high resource allocation, it adjusts parameters such as CPU shares, memory limits, and I/O priorities according to real-time metrics, ensuring adequate performance while optimizing utilization.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The system implements self-service mechanisms where network functions automatically monitor their own resource needs and can request additional resources when required. This allows the system to maintain reliability without permanent over-allocation, as resources are allocated on-demand based on actual usage patterns.

Inventive Principle:

Principle #25Self-service

3Reliability

If resource allocation is dynamically adjusted, then network function performance is maintained, but system complexity increases

Engineering Contradiction:

Improvenetwork function performanceVSAvoidresource management system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system introduces an intermediary layer consisting of resource management controllers and orchestration software that mediate between physical hardware resources and virtual network functions. This intermediary abstracts the complexity of dynamic resource adjustment, providing automated policies and algorithms that handle the intricate coordination without requiring direct complex management at each level.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Reliability

If predictive analytics are implemented for resource allocation, then resource availability is improved, but computational overhead increases

Engineering Contradiction:

Improveresource availabilityVSAvoidcomputational overhead

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system implements partial predictive analytics by applying machine learning models selectively to the most critical network functions and resource types rather than comprehensively to all resources. This partial action approach provides sufficient predictive capability to ensure resource availability for critical functions while limiting the computational overhead to manageable levels.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS10841173B2System and method for determining resources utilization in a virtual network

Publication Date: 2020.11.17 INTUIT INC

AI summary

A system and method for determining optimized resources utilization in a virtual network. The method includes collecting a computing resource parameter from a virtual network, where the computing resource parameter includes at least performance measurements of computing resources in the virtual network; accessing a key performance indicator, where the key performance indicator includes measurement of network traffic performance; and determining, based on the computing resource parameter and the key performance indicator, an optimized resource parameter over a period of time that indicates a usage rule for a computing resource component.