Cloud Workload Analytics Orchestration for Relevant Metric Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In complex cloud and on-premise computing environments, determining and analyzing relevant performance metrics for software components is technically challenging due to the difficulty in identifying deployed workloads and computing resources, which can lead to inefficiencies in monitoring and diagnosing failures.

Innovation Solution

An orchestration system determines and deploys workloads based on network topology models, configures analytics collection and presentation, and selects relevant performance metrics from computing resources to provide accurate analytics dashboards and alerts for software components.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If analytics services collect performance metrics from all computing resources in the environment, then comprehensive monitoring coverage is achieved, but system complexity and data processing overhead increase significantly

Engineering Contradiction:

Improvemonitoring coverageVSAvoidanalytics system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and collects only the specific performance metrics that are relevant to the requested software component's functionality, rather than collecting all possible metrics from all computing resources. This selective extraction approach maintains comprehensive monitoring coverage for the component of interest while reducing overall system complexity and data processing overhead.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The analytics system applies local quality by customizing the set of performance metrics collected based on the specific software component being monitored. Different software components have different functional requirements, so the system tailors the analytics collection to match the local needs of each component rather than applying a uniform monitoring approach across all resources.

Inventive Principle:

Principle #3Local quality

2Loss of information

If the orchestration system collects and presents all available performance metrics, then complete visibility into system performance is achieved, but information relevance and analytical precision decrease due to data overload

Engineering Contradiction:

Improveperformance data completenessVSAvoidmetric relevance

Core Design Contradiction:

Loss of informationVSMeasurement precision

Solution Approach 1:

The system extracts only the performance metrics that are functionally relevant to the requested software component, filtering out unnecessary data. This extraction process prevents information loss for the specific component being monitored while improving measurement precision by eliminating irrelevant metrics from the analysis set.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of starting with all available metrics and filtering down, the system inverts the approach by starting with the functional requirements of the software component and selecting only the metrics needed to fulfill those requirements. This inversion ensures that all collected information is directly relevant to the component's performance.

Inventive Principle:

Principle #13The other way round (Inversion)

3Reliability

If analytics data is collected from multiple workload resource domains with different deployment models, then comprehensive performance monitoring is achieved, but the difficulty of identifying and correlating metrics increases

Engineering Contradiction:

Improveperformance monitoring accuracyVSAvoidmetric identification complexity

Core Design Contradiction:

ReliabilityVSDifficulty of detecting and measuring

Solution Approach 1:

The orchestration system acts as an intermediary that receives the network topology model, identifies the deployed workloads and computing resources, and correlates the performance metrics from multiple workload resource domains based on this information. This intermediary approach maintains monitoring accuracy across hybrid cloud environments while reducing the complexity of metric identification and correlation.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary identification and correlation of computing resources and workloads using the network topology model before collecting performance metrics. This preliminary action establishes the relationships between metrics and components in advance, making the subsequent metric collection and analysis process simpler and more accurate.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12530275B2Orchestration of analytics in workload resource domains

Publication Date: 2026.01.20 CISCO TECHNOLOGY INC

AI summary

Techniques are described for configuring and using analytics for requested software components deployed within cloud-based and other distributed computing environments. An orchestration system may determine particular analytics for a requested component implemented within a computing environment, based on the deployed workloads and/or computing resources on which the requested component was deployed. In conjunction with orchestration of the requested component, the orchestration system may determine the associated performance metrics, including particular telemetry metrics and/or composite metrics, based on the orchestration of the requested component. The orchestration system also may manage the presentation of the performance analytics for the requested component, including customized dashboards with component-specific metrics and/or alerts to target devices associated with the requested component.