Brussels / 1 & 2 February 2020

schedule

Monitoring and Observability devroom


09 10 11 12 13 14 15 16 17 18
Sunday Intro
Monitoring and Observability Devroom
Distributed Tracing for beginners Grafana: Successfully correlate metrics, logs, and traces Jaegertracing in Ceph
An interesting case of distributed tracing
Stories around ModBus
Why ModBus is worse than SNMP
Monitoring strawberries
Building observability for indoor farming
Querying millions to billions of metrics with M3DB's inverted index Secret History of Prometheus Histograms Are You Testing Your Observability? Patterns for Instrumenting Your Services How to measure Linux Performance Wrong From Zero to Useless to Hero: Make Runtime Data Useful in Teams Grafana-As-Code: Fully reproducible Grafana dashboards with Grafonnet Monitoring of a Large-Scale University Network: Lessons Learned and Future Directions

Monitoring and observability are two angles on the same principle: Any computerized system needs to have continuous and computerized verification of its state.
In recent times, the term "monitoring" has sometimes been used for situations in which arbitrary data is being collected and never looked at. This is a misuse, or a misnomer, but it still happens. The term "observability" tries to re-sharpen the underlying concept, trying to emphasize that in the end humans and computers need to be able to observe and grok the data. It also implies that ever-increasing complexity of our modern world needs to be met by automation of at least equal capability to contain said complexity.

The three focus points of this devroom will be general concepts, specific developments and projects, and sharing of experience. We will try to balance content in a way that is beneficial to newcomers and experienced people alike.

As this devroom tends to be always overful, we request that you ideally stay for more than one talk at any one time. Once a talk ends, please leave quietly so we can run our speaker Q&A at the same time. If you want to enter, please wait outside the room not only until people start leaving, but until they are actually done leaving so you can enter more efficiently and quietly. Keep in mind that you will enter during the aforementioned Q&A. Every time people leave and come, we ask people who stayed in the room to defrag into the middle of the rows to free up seats on the row ends.

Event Speakers Start End

Sunday

  Intro
Monitoring and Observability Devroom
Richard Hartmann 09:00 09:05
  Distributed Tracing for beginners Juraci Paixão Kröhling 09:10 09:35
  Grafana: Successfully correlate metrics, logs, and traces Andrej Ocenas 09:50 10:15
  Jaegertracing in Ceph
An interesting case of distributed tracing
Deepika Upadhyay 10:30 10:55
  Stories around ModBus
Why ModBus is worse than SNMP
Richard Hartmann 11:10 11:35
  Monitoring strawberries
Building observability for indoor farming
Jean-Marc Davril 11:50 12:15
  Querying millions to billions of metrics with M3DB's inverted index Rob Skillington 12:30 12:55
  Secret History of Prometheus Histograms Björn Rabenstein (Beorn) 13:10 13:35
  Are You Testing Your Observability? Patterns for Instrumenting Your Services Bartek Plotka, Kemal Akkoyun 13:50 14:15
  How to measure Linux Performance Wrong Peter Zaitsev 14:30 14:55
  From Zero to Useless to Hero: Make Runtime Data Useful in Teams Florian Lautenschlager, Robert Hoffmann 15:10 15:35
  Grafana-As-Code: Fully reproducible Grafana dashboards with Grafonnet Julien Pivotto, Malcolm Holmes 15:50 16:15
  Monitoring of a Large-Scale University Network: Lessons Learned and Future Directions Simone Mainardi, Tobias Appel 16:30 16:55