Linear Predictor for System Metric Threshold Prevention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enterprise systems face challenges in maintaining reliable operation due to potential failures that can disrupt business operations, necessitating improved system management techniques to prevent undesired states such as high response times, low disk space, excessive heap memory usage, and high RAM usage.
Innovation Solution
A method and system that utilize a system manager to monitor system metrics, detect approaching thresholds, calculate predicted values using a linear predictor trained with unevenly sampled data, and reconfigure computing devices to prevent these thresholds from being exceeded, employing remediation actions like restarting processes or moving data to prevent system failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If system metrics are monitored continuously and thresholds are enforced strictly, then system reliability is improved, but system complexity and resource consumption increase
Solution Approach 1:
The system performs preliminary actions by predicting future metric values before thresholds are exceeded. The linear predictor calculates predicted values based on historical data, and remediation actions are triggered proactively when predictions indicate threshold violations, preventing failures before they occur rather than reacting after thresholds are breached
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring system metrics, comparing actual values against predicted values, and automatically triggering remediation actions when deviations are detected. The system manager receives feedback from the predictor and adjusts system configuration accordingly, creating a closed-loop control system that improves reliability through adaptive response
2Reliability
If remediation actions are taken proactively to prevent threshold violations, then system downtime is reduced, but computational resources and processing time increase
Solution Approach 1:
The system applies partial action by using a linear predictor that processes only the essential features of historical metric data rather than analyzing complete system states. The predictor focuses on identifying trends and patterns necessary for threshold prediction, using computational resources selectively rather than performing exhaustive analysis of all system parameters
Solution Approach 2:
The system changes parameters by transforming historical metric data into predicted future values through the linear predictor. The predictor modifies the representation of system state from raw historical measurements to forecasted metric values, enabling proactive threshold detection with reduced computational overhead compared to analyzing complete historical datasets
Data Source
AI summary
A method is disclosed including: obtaining one or more values of a system metric, the system metric being associated with a hardware resource of a computing device; detecting whether the system metric is approaching a threshold, the threshold being associated with a key performance indicator (KPI) of the computing device, the detecting being performed based on the obtained values of the system metric; calculating a predicted value of the system metric in response to detecting that the system metric is approaching the threshold, the predicted value of the system metric being calculated by using a linear predictor that is trained using unevenly-sampled training data; detecting whether the predicted value of the system metric exceeds the threshold; and reconfiguring the computing device to prevent the system metric from reaching the predicted value in response to detecting that the predicted value exceeds the threshold.