Skip to main content


Geecon

GeeCON 2016: Big Data Inspirations

Geecon isn’t only a Java event. It’s also a great place for Big Data and Machine Learning enthusiasts – like me. There were a lot of new technologies most companies have probably never heard of. I’m quite sure they will be using them in the near future, especially if they plan to utilize huge amounts of data or cooperate with thousands of users in the real time. So, what’s cool in the world of Big Data?

During the first day, the best talk (for me) was “Lisp in the Machine” by Joe Nash. He showed how his company, Braintree, deals with billions of events/messages per day, transmitted by Kafka to Redshift and ElasticSearch. So nice!

Now, Machine Learning for JVM developers by Mateusz Dymczyk. I’m not a huge fan of Spark MlLib, mostly because of how difficult it is (you can read what I’ve written about Azure Machine Learning here). Mateusz demonstrated how to easily mark if an email message is spam or not. It was amazing. I never thought this prediction could be so easy to do. I mean, he used just a few lines of code. I think many email providers should use this solution – could you imagine – just by using 50 lines of code there is 90% chance you won’t be getting spam messages, ever?

And last but not least, another good presentation from the first day – Hands of Apache Ni-Fi by Joe Percivall. Every developer should take a look at Ni-Fi. What’s Apache Ni-Fi? Powerful and scalable directed graphs of data routing, transformation, and system mediation logic. Just by dragging and dropping a couple of blocks you can create an entire Big Data ecosystem. At the moment, Ni-Fi supports hundreds of libraries including Kafka, Apache Spark, Cassandra or Parquet. To think that 5 years ago no one had even heard of Big Data. Today, you can use your browser to filter and manipulate terabytes of data, no code needed.

Day two: DuyHai Doan’s two presentations about new features in Cassandra 3.0, and later, an introduction to Apache Zeppelin. Just like last year, he showed where Cassandra is currently going – his presentations are always so satisfying. They’re simple, yet really professional and useful. Take a look right here!

cassandra

Third day – my second rendezvous with Amira Lakhal, Spark and Cassandra – as always, very professional. You can take a look at my first impressions of Amira’s presentation. Going further: AKKA, AKKA, AKKA by Konrad Malawski. As an AKKA evangelist, he demonstrated a very high level of developer skills (as always). I’m wondering when I’ll be as good as he is in AKKA – probably never.

akka

To sum up…

We’re currently highly involved in creating our Big Data Ecosystem (read more here). The presentations I attended at Geecon reassured me that we’re on the right track (and we’re using the right solutions) to design this ecosystem. However, many presentations influenced me to improve it. If you missed Geecon, you should regret this decision! But you can also take a look at http://devoxx.pl/, where many Geecon speakers will be turning up.

Espeo had a proud representation at Geecon, so keep an eye out for our Java crew’s highlights!

sebgeec4

Share:Share on FacebookTweet about this on TwitterShare on LinkedInShare on Google+Pin on Pinterest

  • Mateusz Dymczyk

    Glad you liked the example! You can find more in Spark’s documentation, “Learning Spark” book or Sparkling Water docs 🙂

    I do agree MLlib is a bit hard and more to build machine learning products than for everyday use by devs… That’s why we’re working on something new and amazing at H2O, be on the lookout :-))

    p.s. you do know Konrad is a core Akka DEV? 😉

  • Pingback: Geecon 2016: Java 9, Spring and the Nyan Cat - Espeo Software - Global Software House()

Like what you see?

Get in touch! We'll respond quickly, and we'll keep your data confidential.