ContainerDays 2019

ContainerDays 2019 took place at the Port Museum Hamburg - 2 days of conference talks, 4 parallel tracks, an awesome place, lots of people and high temperatures. On day 1 I attended the following talks:

• How AI and Big Data Platform as a Service work together - a showcase by Dennis Haßlöcher & Christopher Blaschke
• Organizing the YAML mess with Kustomize by Florian Assmus
• Knative by Roland Huß & Matthias Wessendorf
• What does the kubelet say? by Neven Miculinić
• Service Mesh Magic for Managed Kubernetes by Franz Munz

Lunch break: a very good selection of food trucks

Due to the weather conditions I skipped the following talks that I planned to attend:

• Creating a fast Kubernetes Development Workflow by Bastian Hofmann
• Death of the Developer by Karsten Samaschke
• The Silence of the Lambs: Promoting and Inspecting Source Code and Binaries in Continuous Delivery by Michael Hüttermann

These talks I had selected for day 2, but due to clash of appointments and the weather experiences from day 1 I completely skipped day 2 and had a nice and cool ice cream :-)

• Container Isolation Approaches for Multi-Tenant Container Workloads and Serverless Computing by Christian Bargmann (lightning talk)
• Quirks of scaling up your Custom Controllers by Ashutosh Kumar & Amit Das (lightning talk)
• Moving to Kubernetes: the Bad and the Ugly by Maxime Lagresle
• Using ServiceMesh technology in Kubernetes and the integration with legacy services by Stephan Fudeus
• Battle of the Circuit Breakers: Hystrix vs. Istio by Nicolas Frankel
• 7 Deadly Sins of React by Ersel Aker
• Kubernetes Pod: deep dive by Dario Tranchitella
• State of Containers in Big Data and Big Compute by Christian Kniep

The slides are available here. The recordings should be available on the corresponding YouTube channel.

First Meetup in Hamburg

Today was a meetup of the Big Data & NoSQL Hamburg group. The two talks were about Apache Flink, a quite new stream processing system. We've learned about the key features of Flink and got a live demo of its failover and recovery abilities. My takeaways:

• It offers different APIs (for batch and stream processing).
• It can run in different modes incl. a cluster mode on top of YARN.
• Although running in a JVM it implements a memory management on its own to mitigate OutOfMemory failures by serializing objects to disk when heap space gets terse.
• It uses the Chandy-Lamport algorithm for its fault tolerance. Therefore, checkpoints based on time criteria are inserted into the stream to have a well-defined recovery point - in contrast, Apache Spark uses a size criteria to create its mini batches. It is planned to have more control over the segmentation criteria in the future.
• At least the Scala examples look quite close to the ones I know for Apache Spark.
• It still lacks high availability for its job manager (but it's on the roadmap).

The meetup was hosted by Smaato which also offered food and drinks ... plus, their office offers a nice view over Hamburg: