DevOps Porto #37 ONLINE: Monitoring

Welcome to our 37th meetup, now online, about DevOps values, practices and culture.

• Why averages lie by Filipe Oliveira

We spend most of our time looking at the reported averages of our monitoring systems, completely disregarding the painful truth that the numbers that we look at and present to our bosses, to our business and make decisions based upon, do not represent our user experience.
This simple fact seems to surprise many people. It feels good looking at steady state monitoring charts.

Our first invited speaker, Filipe Oliveira, will share why is it important to pay to the "higher end" of the percentile spectrum in most application monitoring, benchmarking, and tuning environments and how you can make better usage of the open-source tooling we have at our disposal.

Filipe is a Performance Engineer at Redis Labs, working on both the enterprise and the open-source products as part of the CTO research group. Before Redis Labs, he worked both on a large scale retail company and as Grid and Distributed Computing Researcher recurring to advanced computing infrastructures listed in the TOP500 supercomputers list.

• Application monitoring and metric collection for Scala By Carlos Teixeira

More often than we would care to admit, we've been stuck looking through endless lines of logs, code, documentation, stackoverflow posts, ... trying to find the reason behind the mysterious disappearance of some row in our database , or why some service always seems to fail for no apparent reason the first Tuesday of each month..

Unfortunately for us, the answer isn't always obvious and usually requires an analysis of multiple services and how they interact with each other. In this case, for instance, maybe a monthly report is generated on the first Tuesday of each month which makes way to many requests to our service causing our db connection pool to throw random errors and, as a consequence, random fails occur when inserting values in our database;

Funny enough, during the entire time, logs were being produced with the exact error but we just weren't able to see them in the middle of all the noise..

In this talk, Carlos Teixeira, will look into how we can integrate and use Kamon to give us a better view of what's going on in our applications and how they interact with each other. He'll look into how to measure cpu usage, query times, individual endpoints performance, custom metrics as well as how to add distributed tracing to our applications so that we can now how they work and how they interact with each other.

After an initial contact with DevOps while still a Uni, Carlos' continued his career as a Backend Software Engineer. Having experienced the value that DevOps can bring to an organization he continued to learn about and explore all things DevOps.

Agenda:
• Welcome
• Why averages lie by Filipe Oliveira
• Application monitoring and metric collection for Scala By Carlos Teixeira
• Closing