Tutorial Big Data Analytics at 3^rd NESUS Winter School

At the occasion of the 3^rd NESUS Winter School and PhD Symposium on Data Science and Heterogeneous Computing, I was invited to give a lecture and hands-on on “Big Data Analytics”. I would like to thank the local organizers (in particular Prof. Karolj Skala and Dr. Davor Davidović), as well as Prof. Jesus Carretero Perez, for this excellent event.

   3^rd NESUS Winter School

• Date: Tuesday January 23th, 2018, 9h – 13h.
• Location: Zagreb, Croatia
• Slides (PDF) – Reference site: http://nesusws-tutorials-bd-dl.rtfd.io/
• Github sources

This tutorial will offer a synthetic view of Big Data Analytics challenges, the tools permitting to address these challenges and focus on one of these tool through a practical session with a set of concrete examples.

Time	Session
09:00 - 09:30	Discover the Hands-on tool: Vagrant
09:30 - 10:00	HPC and Big Data (BD): Architectures and Trends
10:00 - 10:30	Interlude: Software Management in HPC systems
10:30 - 11:00	[Big] Data Management in HPC Environment: Overview and Challenges
11:00 - 11:15	Coffee Break
11:15 - 12:30	Big Data Analytics with Hadoop & Spark
12:30 - 13:00	Deep Learning Analytics with Tensorflow
13:00	Lunch

Title: Big Data Analytics: Overview and Practical Examples

   Online Tutorial: nesusws-tutorials-BD-DL.readthedocs.io

   Download the slides (PDF)

Topics

• Focus on practicals tools rather than theoretical content
• starts with daily data management…
  • … before speaking about Big data management
  • in particular: data transfer (over SSH), data versioning with Git
• continue with classical tools and their usage in HPC
  • review HPC environments and the hands-on environment
    • reviewing Environment Modules and Lmod
    • introducing Vagrant and Easybuild
  • introduction to Big Data processing engines: Hadoop, Spark
  • introduction to Tensorflow, a Machine Learning (ML)/Deep Learning (DL) processing framework

Level: beginner - advanced

Below are a couple of pictures taken during my intervention at the school – you can find all the photos taken in the School photo gallery.

Credits for the pictures: Dr. Davor Davidović