Evolution of a Monitoring System of Computing Resources of the HybriLIT Heterogeneous Platform
摘要
Abstract
This paper presents an overview of monitoring systems for various components of the HybriLIT heterogeneous platform. The goals and purposes of these systems, which serve as important tools for system administration of the platform, are outlined. Previously developed monitoring system is used for monitoring the state of computing resources of the platform. It allows real-time tracking of the load of CPU and GPU of the computing node, RAM usage, storage systems usage, network traffic and etc. This paper presents a new monitoring system which is a logical evolution of the previously developed system. It provides a number of additional functions for monitoring the state of computing resources of the platform.