Telemetry-Based Virtualized Node Selection for Capacity Fit

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In cloud computing environments, accurately predicting whether a virtualized node is overprovisioned or underprovisioned for a given application is difficult due to variations in physical processor architectures, leading to inefficient resource utilization and potential performance degradation.

Innovation Solution

An optimization device determines processing performance scores using a baseline score and relative ratios for virtualized nodes, combined with telemetry data to identify recommended nodes that meet the application's requirements, improving resource utilization efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If virtualized nodes are selected based on default or simplified metrics, then the selection process is fast and simple, but the accuracy of predicting whether a node is overprovisioned or underprovisioned deteriorates

Engineering Contradiction:

Improveaccuracy of predicting overprovisioned/underprovisioned statusVSAvoidcomplexity of selection process

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transforms the node selection process by changing the parameters used for evaluation from simple metrics to a comprehensive scoring system that incorporates processing performance scores, relative ratios, telemetry data, and processing capacity utilization scores. This allows accurate prediction of overprovisioned/underprovisioned status while maintaining automated decision-making

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent introduces an optimization device as an intermediary between the application and virtualized nodes. This device acts as a mediator that performs complex calculations and evaluations, shielding users from the complexity while providing accurate node recommendations based on multiple factors including processing performance and telemetry data

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If comprehensive telemetry data and multiple scoring metrics are used to evaluate virtualized nodes, then the accuracy of node selection improves, but the computational complexity and processing time increase

Engineering Contradiction:

Improveaccuracy of node selectionVSAvoidprocessing time for node evaluation

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by pre-calculating and storing processing performance scores and relative ratios for each virtualized node before actual deployment decisions are needed. This allows the optimization device to quickly retrieve and compare pre-computed metrics rather than performing complex analyses in real-time, reducing processing time while maintaining accuracy

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces manual or simple automated node selection with an optimized computational system that uses structured scoring models and pre-processed telemetry data. This substitution creates an efficient automated decision-making mechanism that handles complexity systematically while providing fast results

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If processing performance scores are calculated using baseline scores and relative ratios for multiple virtualized nodes, then the resource utilization efficiency improves, but the computational overhead increases

Engineering Contradiction:

Improveresource utilization efficiencyVSAvoidcomputational overhead

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent uses copying by creating standardized templates and models for processing performance evaluation. Instead of analyzing each virtualized node from scratch, the system uses baseline processing performance scores and relative ratios that can be replicated and compared across multiple nodes, reducing redundant computational overhead while improving resource allocation efficiency

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20250284515A1Optimizing virtualized node selection

Publication Date: 2025.09.11 CAPITAL ONE SERVICES LLC

AI summary

In some implementations, a device may obtain telemetry data associated with the cloud computing environment, wherein the telemetry data indicates one or more processing parameters of an application deployed in the cloud computing environment via a current virtualized node. The device may determine, for the application, a processing capacity utilization score of the application. The device may determine, based on the processing capacity utilization score and the one or more processing parameters, one or more recommended virtualized nodes from one or more virtualized nodes associated with the cloud computing environment. The device may perform an action associated with a selection of an operating virtualized node for the application from the current virtualized node and the one or more recommended virtualized nodes.