Monitoring huge OpenStack platforms is a challenge -- conventional monitoring tools quickly reach their limits in the context of massively scalable environments. T-Systems uses Prometheus for seamless Monitoring, Alerting and Trending (MAT) of the Open Telekom Cloud (OTC). This presentation explains what the actual challenges are when monitoring OpenStack and how T-Systems has designed a monitoring platform that fits its needs perfectly.
CERN has been running an OpenStack cloud in production since July 2013, across two data centres in Meyrin, Switzerland and Wigner, Hungary. Adopting open source technologies for the computer centre infrastructure has require significant changes, both technically and culturally. The presentation will cover lessons learnt during the deployment and the outlook for the future with the major growth in capacity needed for the future High Luminosity LHC and other CERN experiments.