Join us for talks on the Hadoop ecosystem and Spark

Thanks to Relay Health for hosting our first Atlanta CUG!

Please join us Tuesday, May 19th from 6 - 8 pm ET

Agenda:

6:00 – 6:30 Networking, Intros

6:30 – 7:00 Hadoop ecosystem - Steven Fuller, Sales Engineer

7:30 – 7:30 Spark - Ted Malaska, Principal Solutions Architect

7:30 - 8:00 Q & A

Spark Abstract:

Mindset is so important to development success. Through the years we have had to master concepts like object-oriented development, function programing, inverse of control, and many more. These concepts are more then APIs or something you get from JavaDoc, they are a thought patterns that we have to understand deep in our gut to enable us to build great scalable maintainable software.

Now with the success of Spark and the need for programs to span many many machines in order to solve some of the world's greatest problems, the art of distributed computing and design has again resurfaced to be hot commodity that can take your development career to the next level, but more importantly, can open the door for some really cool and impactful things we can build.

The goal of this talk is to take a Java and/or Scala developers and first introduce them to basic concepts of what Spark is and how it works. Like walking through what and how things like DAGs, RDDs, Transformations, Actions, and Executors are, so that we can build an understanding Spark from and architecture point of view.

Then we will walk through some examples and talk about how one’s mindset must evolve to see the code as compared to Java or Scala code that would run with in a single JVM. We will learn how to think about JVM locality, memory utilization, network/CPU usage, optimizing DAGs pipe lines, and serialization conservation.