TechSummit Berlin
April 12th, 2017

Join us for an enjoyable day of tech insights

Don't miss this chance to hear expert speakers share tips, tricks and ideas on a wide range of tech topics. Join others from the tech community in a day of informal chat and information exchange – all in the really cool surroundings of Berlin's Kulturbrauerei. It’s a great day out and only €25 for a full day pass (including food!). Order your tickets here and now.

Some of the topics our speakers will cover this year are:


Simon Fisher

Simon Fisher

Sr. Solutions Architect,
Chef Software
Björn Rabenstein

Björn Rabenstein

Production Engineer,
SoundCloud Ltd.
Patrik Karisch

Patrik Karisch

Engineer & Operations,
Pixelart GmbH
Moritz Heiber

Moritz Heiber

DevOps Birth Assistant,
Oliver Hardt

Oliver Hardt

Software Engineer,
The New York Times
Terrence Ryan

Terrence Ryan

Developer Advocate,
Amy Nguyen

Amy Nguyen

Software Engineer,
Robert Abraham

Robert Abraham

Head of IT Operations,
Koen Bollen

Koen Bollen

Backend Developer/OPS,
Monica Sarbu

Monica Sarbu

Team Lead - Beats team,
Tudor Golubenco

Tudor Golubenco

Tech Lead - Beats team,


10h00 - 10h15 Opening Notes
10h15 - 10h45

Monica Sarbu & Tudor Golubenco - Elastic
A "beat" of security, with the Elastic Stack

From guessing a weak password to exploiting zero-day vulnerabilities, there are a lot of ways attackers can get into your private networks. So even if you follow all the best security practices, there’s a good chance you will face a security incident sooner or later. The questions are: how quickly will you find out? how will you respond? how will you know which systems are compromised? In this talk, we’ll show you how to use the Elastic Stack, and in particular Beats, to detect security breaches.

10h45 - 11h15

Robert Abraham - Adjust
Feet On The Ground: Scaling Adjust Without the Cloud

Nowadays most SaaS companies put their trust in cloud platforms such as AWS. Often this is enough for most start ups to kick of their product and avoid handling the difficulties of operating your own infrastructure. At Adjust we decided to run our stack on bare metal servers. In this talk I want to cover why we are going that path and how we tackle the common problems of this approach.

11h15 - 11h30 Coffee Break
11h30 - 12h00

Simon Fisher - Chef Software
Application Automation with Chef Habitat

Habitat is an open-source project by Chef that enables you to ship your application, to any platform, along with all the automation you'll need to manage it in production. In this talk, we'll look at creating the habitat plan, building packages, exporting to common container platforms, and running the habitat packages in their own process management supervisor. You'll learn the steps to create repeatable, portable application environments for your applications from a clean room environment to running them in production.

12h00 - 12h30

Patrik Karisch - Pixelart GmbH
Application load testing with open source and the cloud

Once you created your new website or webapp you're pretty happy to finally release it into production. But do you know if you can handle the load if your site is mentioned on Hacker News or TechCrunch? Did you forget to load test your site? Or are you in pre-production and looking for load testing before going live?

This talk is showing you all the open source tools to load test your site, shows you useful scenarios and gives you introductions to utilize the load from different clouds. We also take a trip and look how you profile your site under load (for example with Blackfire). And last but not least, I will show you how parts of this can be automated with Jenkins.

Whether you have an OpenStack cloud at your hands (awesome, more Open Source \m/) or you're using Amazon or Google cloud. All are covered.

12h30 - 13h30 Lunch
13h30 - 14h00

Björn Rabenstein - SoundCloud Ltd.
Why your alerting should be based on labeled time series

Prometheus has brought many paradigm shifts to the world of open source monitoring and alerting. Perhaps the two most crucial ones will be presented in this talk: First, alerting should be based on time series. And second, everything should be labeled. Both together are crucial for meaningful, actionable alerts when dealing with large scale systems. This talk is an introduction to Prometheus with a focus on these two aspects.

14h00 - 14h30

Oliver Hardt - The New York Times
Monitoring Performance, Reliability and Resource Utilization with Prometheus

The Personalization and Messaging team at The New York Times successfully uses Prometheus ( to monitor a fleet of microservices as well as an older legacy PHP system.

This talk will give a general overview of why monitoring is important, what should be monitored and how Prometheus is a good fit for monitoring modern cloud-native distributed systems. There will also be two real-life case studies, the first one will go into detail how we monitor our Golang and Python microservices, leveraging Consul for service discovery and include examples for what instrumenting Golang and Python code looks like and what metrics can be obtained. The second use case will show how we instrumented a legacy PHP app and were able to extract useful metrics about host and application performance that allowed us to confidently reduce provisioned AWS resources and save about 30% of our cloud cost while maintaining a comparable level of service.

14h30 - 15h00

Amy Nguyen - Pinterest
UX Design and Education for Effective Monitoring Tools

I, like many of us, chose to work in infrastructure because I thought it meant I could avoid talking to other people as much as possible. I was wrong. It turns out that a huge amount of work in monitoring is about empowering the rest of your engineering organization to use the tools you develop correctly, quickly, and effectively. We spend so much time explaining how to interpret timeseries data and why averaging percentiles is a bad idea. After feedback from our engineers, we embarked on a journey to redesign our internal monitoring tools and understand where people were struggling with the existing system.

In this talk, I will explain how we approached the problem of making concepts like interpolation, aggregation, and alerting more intuitive and how we identified pain points for new users. I will outline common misconceptions our users have about monitoring and how we cleared up this confusion in our UI without forcing everyone to spend hours on documentation. Rather than copying and pasting existing UX design principles onto our monitoring problems, we will see how we can reinterpret these ideas and apply them to our unique situation to create a better experience for everyone.

15h00 - 15h15 Coffee break
15h15 - 15h45

Terrence Ryan - Google
LAMP in Containers

There is momentum to moving container based systems instead of the traditional single machine (or virtual machine) model. Where does that leave your mission critical LAMP stack application? Can it be containerized? Without tremendous amounts of pain? Of course it can.

Containerizing LAMP stack apps is pretty straight forward, once you know what you're doing. The goal of this session is to get you there. This session will walk you through setting up a LAMP app on an existing Kubernetes cluster using Replica Sets, Stateful Sets, Persistent Volume Mounts, and other Kubernetes tools you need to run the LAMP stack in a container based system.

15h45 - 16h15

Moritz Heiber - Thoughtworks
Playing Port Authority - TDD for containers

I will talk about unit-, functional- and integration testing when developing container based infrastructure, using ServerSpec, Ruby and docker-compose intertwined with Continuous Delivery. This will include a live coding example as well as a step by step introduction and ready-to-use sample code. The title is aimed at "playing port authority" in a sense that you're inspecting your containers before shipping them "somewhere".

16h15 - 16h45

Koen Bollen - Blendle
Kubernetes in production since 0.9. What we learned, gained, lost and look forward to.

Since January 2017 Blendle has moved almost entirely from GCE VMs to Kubernetes, VMs which were migrated from AWS and initially hosted on bare metal at LeaseWeb. 403 pods, 216 deployments, 107 services, 41 ingresses on 9 nodes. This talk is a summary of our adventure moving to Kubernetes in production and an insight in how we decided to use it. We'll take a look at issues we've encountered, talk about some pros and cons, and see what our next steps are going to be.

16h45 - 17h00 Closing Notes
17h00- 18h00 Drinks


Please note that by registering you agree to our Code of Conduct

Ticket sales via Eventbrite


Kulturbrauerei, Berlin

The KulturBrauerei is a former landmark of industrialization with a distinctive architecture from the end of the 19th century – located in the heart of Berlin. The site which is under national trust protection contains six linked courtyards and over 20 buildings used by creative artists for a wide range of events including cultural happenings like parties, theatrical performances or cinematic and gastronomic events.

Kulturbrauerei, Berlin



Hosted by:

Click here for more information about sponsorship opportunities

Media Partners: