Apache Flink Coding Dojo

Hope everyone had a great summer. It's about time to start streaming again!

September's meetup will take a different format compared to our previous events. Based on your good feedback about having hands-on exercises, we decided to have a Flink coding dojo.

For those of you who have never been to one, a coding dojo is an event where people work in small groups for a short period of time (2 - 2.5 hours in our case) on topics of their choice and later share their achievements with the rest. Or in other words, it's a perfect chance for you to start using Flink. We'll make sure that each group has at least one coder who already knows the API.

I will give more details and topic suggestions when I get back from holidays in a couple of weeks! :-)

In the mean time, please sign up. See you all soon.

Reminder: To get into the venue you must register with SkillsMatter https://skillsmatter.com/meetups/8079-apache-flink-september-meetup

UPDATE

Here are some topic suggestions:

• Python API

There's a beta version of a Python API (https://github.com/apache/flink/tree/master/flink-libraries/flink-python) (at the moment, only batch API is available). A good way to explore it would be to re-implement one of the existing Scala and Java examples in Python (suggestion: k-means clustering example (https://ci.apache.org/projects/flink/flink-docs-release-0.7/run_example_quickstart.html), which was presented in the first meetup). Useful resources:

• Getting started (http://www.willmcginnis.com/2015/11/08/getting-started-with-python-and-apache-flink/)

• Getting more complicated (http://www.willmcginnis.com/2015/12/13/pyflink-getting-a-bit-more-complex/)

• Python API example (https://github.com/wdm0006/flink-python-examples)s

• Integration with other technologies

Implementing a full streaming pipeline would be useful for those who want to use Flink in practice. Suggestion: Kafka => Flink => write to DB. Useful resources:

• Kafka connector (https://ci.apache.org/projects/flink/flink-docs-release-1.1/apis/streaming/connectors/kafka.html) & how-to walkthrough (https://www.youtube.com/watch?v=7RPQUsy4qOM)

• For beginners: getting to know the DataStream API

Simply ingesting a data stream, doing some filtering/transformations, and outputting the result. For example, ingest a tweet feed => track the number of tweets with some hashtag during the last 15min => output and visualise the trend. Useful resources:

• Programming guide (https://ci.apache.org/projects/flink/flink-docs-release-1.1/apis/streaming/index.html)

• Demo (https://github.com/dataArtisans/flink-streaming-demo)

Note that these topics are only examples and everyone is highly encouraged to propose their own.