Jörg is a software engineer at Mesosphere in Hamburg. In his previous life he implemented distributed and in memory database and conducted research in the Hadoop and Cloud area. His speaking experience includes various Meetups, international conferences, and lecture halls.
The application landscape inside our datacenter is changing: there are a number new distributed data processing frameworks such as Kafka or Flink being released on a weekly basis and also the trend towards microservices and container. This has implications for the ways we are running our datacenter. With this growing need of computing power, distributed applications, and larger data centers also the need for a reliable and simple use cluster manager and programming abstraction grows. This talk explains how Apache Mesos and DC/OS allows combining Microservice management and Fast Data systems on a single platform.
The current craze of Docker has everyone sticking their processes inside a container… but do you really understand cgroups and how they work? Do you understand the difference between CPU Sets and CPU Shares? Spark is a Scala application that lives inside a Java Runtime, do you understand the consequence of what impact the cgroup constraints have on the JRE? This talk starts with a deep understand of Java’s memory management and GC characteristics and how JRE characteristics change based on core count. We will continue the talk looking at containers and how resource isolation works. The session will detail specifically the difference between CPU sets and CPU shares and memory management. The session will close with a deep understanding of the consequences of running the JRE in a CPU share environment and the potential for pseudo-random behavior of running in a heterogeneous datacenter.