Vijay Samuel

We are used to calling Metrics, Logs, Events, Traces as the 4 pillars of Observability. True Observability is not attainable without being able to jump from one pillar to another. Exemplars promote the same from Metrics to Traces. Over the years, we have found extreme value in being able to use Exemplars for other things such as making tail sampling decisions, enriching alerts with sample traces and finally serving longer term Exemplars out of our tracestore built on top of ClickHouse. In this talk, we discuss how metrics and traces go hand in hand through the concept of Exemplars.

OpenTelemetry is today's defacto standard for distributed tracing. It's collector is used by everyone in varied configurations to accept spans. However, tuning the collector takes time and a lot of energy to get right. In this talk we discuss what are the steps we took to tune the collector, what our final configuration was and how we pivoted the platform to a more optimal architecture based on what we learned. 

This talk provides a reference implementation for tracing at scale. Something we don't hear enough about.

Running a metrics platform at scale is not an easy task. Systems like Prometheus offer cloud native monitoring but have their vertical limits. Sharding and clustering allow us to spread the problem out but it isn't infinte. Decentralizing the TSBDs and federating the queries in a smart way is the way to go. In this talk we discover how exactly to do that, what the benefits of going planet scale are and what are the key takeaways from our journey.