Percona PMM: Integrate Percona Cluster Monitoring within an existing monitoring platform
Percona Monitoring and Management is a very powerful set of tools. Built on top of prometheus and grafana, it comes bundled with major parts of a wide range monitoring system (not only database). As most of IT systems already have a monitoring platform, How to integrate PMM within an existing system ? what kind of customization should be made to ease the process ? How to stay as close as possible to the upstream project ? Can we avoid having duplicated monitoring features? In this demo, i'll show you an example of Percona PMM integrated within an existing monitoring platform and try to answer most of these questions. The monitoring system is mostly built on top of collectd and graphite, but the overload idea could be applied to nearly any modular monitoring system. We'll use tools from the Docker ecosystem to ease the integration process and put some stress on a Percona Cluster deployed in a Swarm cluster using Stress tools like tpcc and/or sysbench, to check how the overall monitoring system behave with PMM as an addon.
Rachid is a former virtualization consultant and Instructor. After a successfull experience building and training the ops team of the french registry (AFNIC), he's now C.I.O of a worldwide recognized CRM and ecommerce agency. Dedicated to offer the most efficient production tools and infrastructure platforms, helping developpers focusing on their project, he works with his team on bringing latest technology at a production level. Involved in many OSS project from monitoring to containerization solutions. Teaching in a software engineering school, where he is specialized in infrastructure courses and brings devops culture to his student.