Anomaly Detection for Software-Hardware Problem Differentiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software providers face challenges in diagnosing issues with software applications hosted in Internet Area Networks (IANs) provided by third-party providers, as they have limited information about the hardware resources and cannot distinguish between software and hardware-related problems.
Innovation Solution
A technique using unsupervised machine learning for anomaly detection, representing the software application as a topology of nodes and capturing various metrics to identify anomalies, allowing the software provider to determine if issues are related to the software or the uncontrolled platform.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If software applications are hosted in an Internet Area Network (IAN) provided by a third party, then cost is reduced and resource efficiency is improved, but the ability to diagnose software problems is worsened due to limited information about hardware resources
Solution Approach 1:
The patent introduces an intermediary system that collects hardware resource information from the IAN provider and makes it available to the software provider. This intermediary layer bridges the information gap between the software provider and the third-party hardware infrastructure, enabling diagnosis capabilities without requiring direct access to or ownership of the underlying hardware.
Solution Approach 2:
The patent segments the monitoring and diagnosis functionality into separate components: hardware resource monitoring, software performance tracking, and anomaly detection. This segmentation allows the software provider to focus on software-specific metrics while obtaining hardware information through structured interfaces, maintaining diagnostic capability in a cloud environment.
2Productivity
If hardware resources are dynamically assigned on an as-needed basis, then cost is reduced and resource utilization is improved, but the ability to determine the root cause of software problems is worsened
Solution Approach 1:
The patent implements feedback mechanisms that continuously collect and analyze both hardware resource metrics and software performance data. By correlating hardware state changes with software anomalies in real-time, the system can trace problems back to their root causes even in dynamically assigned resource environments, maintaining measurement precision despite resource fluidity.
Solution Approach 2:
The patent creates detailed copies or representations of hardware resource states and assigns them to corresponding software instances. This virtual mapping allows the system to track which hardware resources were assigned to which software applications at specific times, enabling root cause analysis even when physical hardware changes dynamically.
3Ease of operation
If the third-party provider owns and operates the hardware infrastructure, then the software provider avoids infrastructure management overhead, but information about hardware resources becomes limited
Solution Approach 1:
The patent develops a universal information gathering system that can obtain hardware resource information through multiple channels and formats. This multi-functional approach allows the software provider to maintain ease of operation by not managing infrastructure directly, while still accessing comprehensive hardware information through standardized interfaces and APIs that work across different IAN providers.
Data Source
AI summary
A computing platform may obtain observed data vectors related to the operation of a topology of nodes that represents a software application running on an uncontrolled platform, wherein each observed data vector comprises data values captured for a given set of operating variables at a particular point in time. After obtaining the observed data vectors, the computing platform may apply an anomaly detection model to the observed data vectors and then based on the anomaly detection model, may identify an anomaly in at least one operating variable. In turn, the computing platform may determine whether each identified anomaly is indicative of a problem related to the application, and based on a determination that an identified anomaly is indicative of a problem related to the software application, cause a client station to present a notification.