Last month the Weavework’s team headed to Munich, Germany for the second annual PromCon – a fantastic conference around Prometheus, the open source time series database.

The Google office in Munich was an excellent venue, with a large auditorium filled to the rim with roughly 200 enthusiastic attendees.

Talks and presentations 

Julius Volz kicked off the agenda with a quick recap of the past 12 months - an impressive growth graph of Github Stars and < 200% growth of contributions and third party integrations.

Next up was Matt Bostock from Cloudflare sharing insights and best practices on monitoring a planet-scale edge network with Prometheus. They have 185 Prometheus servers in production worldwide, which caused many “Oooohs” from the attending crowd. Here is the visualization of Cloudflare’s Routing Tree:

For more learnings, take a look at Matt’s slides and keep an eye on the Prometheus youtube channel for the talk recordings.

One of our personal highlights was of course a deep dive into the past 12 months of Cortex, the basis of our monitoring and analytics capabilities in Weave Cloud.

Tom Wilkie walked through a handful of issues (e.g. such as DynamoDB Write Throughputs, long tail queries and best practices for recording rules and alerts that helped us evolve Cortex’ architecture to the current state.


Take a look at how we run Prometheus-as-a-service, our monitoring best practices or DYI Grafana dashboards

Who won the Lego MindStorms?

Many of you entered our prize draw, but there could only be one winner - that lucky person was Kamila Součková (@AnotherKamila).

This was her reaction....

Congratulations Kamila, we can’t wait to see what you come up with!

Until next year

The Prometheus and Google Munich teams did an almighty good job in hosting us, fostering collaboration and allowing for great conversations in the community all day and night.

Take a look also at the wrap up blog and video over on the Prometheus blog.

Thanks again for the Gemütlichkeit and we can’t wait for next year!