Workshops @ Berlin Buzzwords

Right after Berlin Buzzwords 2012, there will be several workshops on related topics, which are available to everyone who has a ticket for Berlin Buzzwords. Those who hold a workshop will provide all the details, number of participants and registration procedure in the Berlin Buzzwords Wiki (that will be online in April).

If you would like to hold a workshop and need assistance with finding a venue or other organizational things, we will support you. Contact us via info@berlinbuzzwords.de and tell us where you need help. Please register your workshop until April 13.

As an example, here's what will happen in the first workshop that is already confirmed:

Topic: 'Parallel Processing beyond MapReduce'

This workshop will discuss novel paradigms for parallel processing beyond the traditional MapReduce paradigm offered by Apache Hadoop.

The workshop will introduce two new systems:

Apache Giraph aims at processing large graphs, runs on standard Hadoop infrastructure and is a loose port of Google's Pregel system. Giraph follows the bulk-synchronous parallel model relative to graphs where vertices can send messages to other vertices during a given superstep.

Stratosphere (https://www.stratosphere.eu) is a system that is developed in a joint research project by Technische Universität Berlin, Humboldt Universität zu Berlin and the Hasso-Plattner-Institut in Potsdam. It is a database inspired, large-scale data processor based on concepts of robust and adaptive execution. Stratosphere offers the PACT programming model that extends the MapReduce programming model with additional second order functions. As execution platform it uses the Nephele system, a massively parallel data flow engine which is also researched and developed in the project.

Attendees will hear about the new possibilities of Hadoop's NextGen MapReduce architecture (YARN) and get a detailed introduction to the

Apache Giraph and Stratosphere systems. After that there will be plenty of time for questions, discussions and diving into source code.

As a prerequisite, attendees have to bring a notebook with:

 - a copy of Giraph downloaded with source

 - Hadoop 0.23+ source tree and JARS local

 - a copy of Stratosphere with source

 - an IDE of their choice

The workshop will take place on the 6th and 7th of June and is limited to 15 attendees. Please register by sending an email to sebastian.schelter @ tu-berlin.de.