Yarn vs Mesos Deathmatch & Scalding over Hadoop

18:00 -18:30 - mingling

18:30 -19:30- Avishai Ish-Sahlom (Infrastructure expert and systems architect @ fewbytes) - Yarn vs Mesos Deathmatch - (first session high level explaination) In recent years hadoop has evolved to become an ecosystem of various data processing projects. As part of this evolution is has attained a grid component - Yarn. However, Yarn isn't the only grid that can run data oriented jobs and many people are interested in Mesos.
This session will give an overview of grids in general, Yarn and Mesos in particular and compare them in the context of data oriented workloads . The things that we are going to focus on Yarn and Mesos and the difference about them.
Yarn vs Mesos Deathmatch has two session
for the second session : Yarn vs Mesos Deathmatch - second session (https://www.meetup.com/Big-things-are-happening-here/events/223599573/)

19:30-19:40 Coffee Break

19:40 - 20:40 Alex Zhicharevich (Research Engineer @ eBay)- Hadoop in
Scala -
Efficient MapReduce with Scalding Scala is a new generation JVM language mixing both OO and functional paradigms.
It's powerful collections api, native support for lambdas, immutable data structures and concise syntax made it a perfect fit for the big data community.
Scalding is a library that enables working with distributed hadoop datasets as natural as working with scala collections.
In this talk I will give a quick overview on Scalding and how it compares to other MR frameworks. I will also present Algebird, a powerful aggregation library that further simplifies some non trivial hadoop analytics.