This thesis is devoted to design environment for automatic horizontal scaling, which will, based on the metrics, that are provided by our environment and saved into database, respond with creation or removal of virtual machines. Environment also has graphs representing resource usages (memory usage, processor usage, numbers of network packets sent, ...), on which we can monitor our system. Environment is built with the help of open source tools OpenStack, Docker, Grafana, Prometheus, Alertmanager, cAdvisor, Node exporter, Vagrant and Ansible.
|