Polygraph-Based Anomaly Detection in Datacenter Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data analytics platforms face challenges in efficiently detecting anomalies and modeling behaviors within cloud environments, particularly in large-scale datacenter settings, where identifying deviations from normal behavior is complex due to the dynamic nature of virtualized resources and ephemeral connections.
Innovation Solution
A data platform is configured to collect and analyze data from agents deployed across compute assets, using polygraphs to model behavioral relationships and detect anomalies by constructing logical graphs of entities and their interactions, enabling real-time anomaly detection and reporting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data analytics platforms collect and analyze data from agents deployed across compute assets in large-scale datacenter settings, then anomaly detection capability is improved, but data size and complexity increase
Solution Approach 1:
The patent segments data into polygraphs that group related entities and their behavioral relationships. By organizing data into these structured graph units, the system reduces overall data complexity while maintaining comprehensive anomaly detection coverage across distributed compute assets.
Solution Approach 2:
The patent introduces polygraphs as intermediary structures between raw agent data and anomaly detection algorithms. These polygraphs serve as a mediating layer that pre-processes and organizes data, reducing the computational burden on analysis systems while preserving essential behavioral patterns for anomaly detection.
2Measurement precision
If data analytics platforms model behavioral relationships using polygraphs to detect deviations from normal behavior, then anomaly detection precision is improved, but device complexity increases
Solution Approach 1:
The patent performs preliminary actions by pre-building polygraphs that capture typical behavioral relationships between entities before anomaly detection is needed. This pre-modeling of normal behavior patterns enables precise anomaly detection without requiring complex real-time analysis, as deviations from pre-established polygraph patterns can be identified more efficiently.
Data Source
AI summary
Detecting deviations from typical user behavior, including: identifying a geographic location of a device that is associated with a user; determining device activity associated with the user; and detecting, based on a profile associated with the user, that the device activity associated with the user deviates from normal activity for the user.