| dc.description.abstract | Nowadays, with the rapid evolution of cloud-native technologies, Kubernetes (K8s) has
been adopted widely as a leading container orchestration platform. In companies, K8s is built
to serve large systems which release many applications on it, with countless pods deployed on
a K8s cluster cannot avoid errors and bugs during operation. Therefore, proficiently operating
a reliable monitoring system for K8s clusters is essential considered when working with Kubernetes to always be able to monitor, warn and detect appropriately the health status of applications in the distributed system.
The main purpose of this thesis focuses on the development and implementation of a monitoring system for applications deployed on Kubernetes using Docker containers, Prometheus,
Grafana and Node Exporter technologies for resource monitoring and performance evaluation.
This was necessary to address server's challenges such as ensuring easy integration, scalability,
high availability, resource management, flexibility, and management. Using visualization techniques to monitor overall usage of resource, energy consumption, and analyze the system's performance on server applications. Furthermore, this system not only allows easy monitoring of
the resource usage status of applications but also provides alerts for administrators and users as
part of its security strategy.
In this thesis, the Prometheus-Grafana tech stack integrated plausible open-source solutions
available on the market. This tech stack is a comprehensive toolset used in Kubernetes to monitor and visualize the K8s ecosystem. As a result of this implementation, a solution was built
and deployed in the K8s production environment, serving as a core system for monitoring and
alerting on K8s service resource usage. | en_US |
