Optimizing Linux Systems: Leveraging Guider for Advanced Performance Analysis and... - Peace Lee

Optimizing Linux Systems: Leveraging Guider for Advanced Performance Analysis and Monitoring - Peace Lee, Hyundai Motor Company

In this presentation, I'll explore Guider, an all-encompassing performance analysis tool, focusing on its automatic performance monitoring capabilities and how to leverage the generated data. The automatic performance monitoring continually observes resource usage and events of systems and tasks, generating performance reports based on collected information when issues arise. All these actions are automatically performed based on various predefined threshold values stored in files. This feature allows for easy identification and analysis of potential performance issues that may be challenging to reproduce or go unnoticed by users. During the operation of the automatic performance monitoring feature, various system and performance information is periodically recorded in a specific file. Representative data generated includes resource usage of systems and tasks, event occurrence counts, peak data, snapshot data, and the number of threshold-exceeding events. Analyzing this data, separated by software versions or hardware types, proves useful in understanding performance differences and changes. The repository for Guider can be found at https://github.com/iipeace/guider.