Programme

Wibauthuis, Wibautstraat 3b, 1091 GH Amsterdam

Choose 1 of 4 tracks

1. Compute: Introduction to UNIX, Cluster computing, HPC Cloud
2. Big Data Analytics
3. Scientific Visualisation and Virtual Reality
4. Sensors and Clouds: program your sensor-node and analyse the data in the cloud - this track is CANCELLED due to a limited number of registrations
5. Artificial Intelligence and Machine Learning

SIGN UP NOW

08:45 - 09:15 | Registration

08:45 - 09:15

Software installation support desk

09:15 - 10:15 | Welcome & Keynote

09:15 - 10:15

Geleyn Meijer, Dean Faculty Digital Media and Creative Industry, HvA 

Nanda Piersma, Professor Urban Analytics (HvA, CWI): 'Big Data in the City of Amsterdam'.

10:15 - 17:00 | Track 1 - Compute: Introduction to UNIX, Cluster computing, HPC Cloud

10:15 - 11:30

Introduction to UNIX

Jeroen Engelberts, Lykle Voort - SURF Supercomputing

Description
You will get acquainted with some basic Unix commands you will need to start using data, compute, and network facilities.

Objectives
At the end of the workshop, you are able to:

  • work with a UNIX terminal/shell
  • use some basic UNIX commands
  • login to a UNIX cluster

Prerequisites
Some basic knowledge working with a computer.

Installation of software needed: Bring along your own laptop. Please install the following tools on your laptop, depending on your operating system:

Before the day starts, you will receive a username and password to login to a UNIX system (in this case Lisa compute cluster and national supercomputer Cartesius). Besides you will create a public/private keypair needed during the HPC Cloud workshop.

12:30 - 14:30

Introduction to Cluster Computing

Jeroen Engelberts, Lykle Voort - SURF Supercomputing

Description
In this part, you will learn how the national cluster Lisa and the national supercomputer Cartesius are set up. This presentation will be followed‐up by a hands‐on with some small and easy to follow examples on one of these systems.

Objectives
At the end of the workshop, you are able to:

  • login to a UNIX cluster
  • prepare, submit and analyze a batch job on the national cluster Lisa / supercomputer Cartesius

Prerequisites
Some basic knowledge of the UNIX operating system.

Installation of software needed: Bring along your own laptop. Please install the following tools on your laptop, depending on your operating system:

Before the day starts, you will receive a username and password to login to Lisa and Cartesius via email.

15:00 - 17:00

Introduction to HPC Cloud Computing

Ander Astudillo, Nuno Ferreira - SURF HPC Cloud

Description
Computing in the cloud allows you flexible and easy access to computing and data resources that you would otherwise have to host yourself. SURFsara runs the HPC Cloud providing an “Infrastructure as a Service” (IaaS) model (as will be explained in the workshop “Introduction to UNIX”). This workshop provides a general introduction to cloud computing, teaches HPC Cloud characteristics and how to use it hands‐on.

Objectives
At the end of the workshop, your are able to:

  • use the HPC Cloud
  • understand and apply different scaling models for parallel computing build (clusters of) Virtual Machines

Prerequisites
This workshop assumes familiarity with the Unix command line and SSH.

Installation of software needed: Bring along your own laptop. Please install the following tools on your laptop, depending on your operating system:

You will need the public/private keypair created during the Introduction to UNIX session.

10:15 - 17:00 | Track 2 - Big Data Analytics

10:15 - 11:30

Big Data - part 1

Machiel Jansen, Mathijs Kattenberg, Matthijs Moed - SURF Scalable Data Analytics

Description
You are introduced to the Spark frameworks for processing big data. These frameworks offer a novel way for creating data analysis applications that easily scale over hundreds to thousands of machines. This data‐parallel approach has been pioneered in industry by tech companies such as Google and Facebook, and is very applicable to many scientific workloads in general. We introduce you to the key concepts and features of the Apache Hadoop and Spark stacks. In addition, you will work on hands‐on Spark exercises in a Jupyter notebook environment. The presentations, exercises and demos will provide a basic understanding of Spark and teach you about fundamental concepts in big data processing.

All three Big Data sessions are related to eachother and should be followed as one package.  

Objectives
At the end of the workshop, you are able to:

  • Understand Spark concepts and fundamentals
  • Understand requirements for scalable applications
  • Run and create basic Spark code in a notebook environment

Prerequisites
This workshop is for anyone who would like to get started with Apache Spark to build robust and scalable applications. You should be familiar with the basics of programming (preferably Python). Most scientific programmers and technically minded researchers will feel right at home.

Installation of software needed
No specific software has to be installed. Bring your own laptop.

12:30 - 14:30

Big Data - part 2

Continuation of part 1

15:00 - 17:00

Big Data - part 3

Continuation of part 2

10:15 - 17:00 | Track 3 - Scientific Visualisation and Virtual Reality

10:15 - 11:30

Introduction Visualisation

First presentation

Mirjam Vosmeer, Researcher Videogames & Virtual Reality (HvA): 
"Are you experienced? Interactive storytelling for VR" 

Is VR a hype? Most certainly. In every possible field people seem to reach out to VR to sell, educate, inform and persuade. However, the exact ways in which VR works or how ‘presence’ impacts an individual user, for instance, are still quite unclear. In our research, we work together with students and industry partners to set up projects in which we investigate the ways VR can be used for storytelling.

At the end of the presentation you will have an idea how complicated - but also how fascinating - the use of VR is.

Second presentation

Sabine Niederer, Professor Visual Methodologies (HvA):
"Introducing Visual Methodologies"

This introduction gives an overview of various analytical practices that use and produce images for research, ranging from so-called “issue mappings” that capture the substance of an online debate, to participatory research that includes annotated maps and visual surveys. The talk also showcases different visual formats for data visualisation and data-driven storytelling, and points to some useful tools so you can start including visualisation in your own research practice. 

At the end of the presentation you gain a basic understanding of how you can use visualisation in your research, not as a final result or illustration of your analysis but as an important part of the research process.

12:30 - 14:30

Scientific visualisation

Paul Melis - SURF Visualisation

Description
A general introduction to data visualisation and its applications in science and research. The session takes a very hands-on approach and is based on freely available (open-source) tools.

Scientific visualisation is about creating representations of 2D or 3D data, which give insight in complex data sets and enable visual exploration and understanding of complex phenomena. We will provide sample data sets and exercises to introduce you to different methods of scientific visualisation such as stream lines, surface rendering and volume rendering.

Objectives
At the end of the workshop, you are able to:

  • Understand how to use scientific visualisation to explore and understand complex data
  • Use open source visualisation tools to make 2D/3D visualisations
  • Have a basic understanding of the visualisation infrastructure available at SURF

Prerequisites
A laptop is required and basic knowledge of Python is preferred but not obligatory.
Required software: Paraview (http://www.paraview.org/). This can also be installed during the workshop.

15:00 - 17:00

Information visualisation

Casper van Leeuwen - SURF Visualisation

Description
The session takes a very hands-on approach and is based on freely available (open-source) tools.

You will learn how to visualize abstract data where no spatial representation is given, such as Google stock data, cluster data, network-graph data and geo-tagged data. It’s an interactive hands-on within a Jupyter notebook environment where you will learn the basics of information visualisation with python frameworks such as Pandas, Networkx and Folium. You will learn how to explore the data and extract the underlying data patterns by applying the right visualisation techniques and using the right data dimensions for the visualisation parameters. Besides the python visualisation you will get a quick glance on some interactive JavaScript d3js visualisations within the Jupyter notebooks.

Objectives
At the end of the workshop, you are able to:

  • Understand why and when to apply basic information visualisation techniques for a given type of abstract data
  • Use open source visualisation tools to make 2D/3D visualisations
  • Use various information visualisation frameworks
  • Have a basic understanding of the visualisation infrastructure available at SURF

Prerequisites
A laptop is required and basic knowledge of Python is preferred but not needed.
Required software: Paraview (http://www.paraview.org/). This can also be installed during the workshop.

10:15 - 17:00 | Track 4 - CANCELLED Sensors and Clouds: program your sensor-node and analyse the data in the cloud

10:15 - 11:30

Introduction Sensors and Clouds

This track is cancelled due to a limited number of registrations.

First presentation 

Marcel van der Horst (Sensorlab HvA): 
From Basic Sensors to Smart Sensor Systems

An introduction to sensors, ranging from physical working principles to self-calibrating smart sensors. Some examples of simple sensors and their application will be discussed. It is intended as an introductory lecture.

At the end of the presentation, you will:

  • know what sensors and sensor systems are
  • you are able to select simple sensor (systems)
  • know their differences, and ways to apply them correctly.   

Second presentation

Mark Borst, Paul 't Hoen and David Quainoo (Big Data group Facility Services UvA/HvA):
On the road: from reaction to proaction. Discovering the use of sensor date for facility management

The facility department (FS) of the UvA/HvA recognizes the importance of big data. As a provider of services to the faculties, FS is aware that if sensor data is used and ‘connected’ properly, it can help to optimize our services. From reaction to proaction.
We present the objectives for the use of big data for FS and show the challenges and results of a real-time use case that we currently do together with researchers from the HvA.

At the end of the presentation, you will: 

  • Learn about tangible possibilities of Wi-Fi data and the physical world
  • You have insight in the real estate and customer objectives of a large organizations

Third presentation

Paul Dekkers (SURF Open Innovation Lab):
Internet of Things, low-power and long-range

Quite a few things in life might be better if they were online: just take a walk through the city and you will see full waste containers, half submerged boats, parking spaces, bicycles, lantern posts. In this presentation preceding the hands-on workshop in session two, we show some appealing examples of the popular Internet of Things (IoT) and you learn the insights of the technique behind it.

At the end of this presentation you will:

  • Know the position of low power wide area networks (LPWAN) in the IoT landscape
  • The function of the protocols needed for IoT
  • Get inspired to build your own sensor

 

12:30 - 14:30

Build your own sensor

Paul Dekkers - SURF Open Innovation Lab 

Build your own sensor

Using a new technology is the best way to familiarize yourself with its features and possibilities: simply playing around with the devices, the network components….

During the workshop we build a temperature and humidity sensor with the Sodaq Autonomo, a matchbox sized, microcontroller board powered by a small solar panel.

The Autonomo board can be programmed with the open source Arduino software and can record data and trigger actions in any environment. It allows you to easily connect (different) sensors.

 

At the end of the workshop, you should be able to:

  • Build your own sensor
  • Know how to receive your data
  • familiarize yourself with the hardware, learn, and have fun

Prerequisites:

You will work with easy to use hardware. Also the copy-paste programmer can easily work with it.

15:00 - 17:00

Analyzing sensor data on the SURFsara IoT platform

Jeroen Schot - SURF Open Innovation Lab

Working with sensor data has its own challenges compared to working with static data-sets. During this workshop we will look at these challenges and show how to tackle them. As an example we will use the platform SURFsara is developing as part of the SURF Open Innovation Lab, but the principles and approaches are also valid for other cloud platforms. The majority of the workshop will be in the form of a hands-on tutorial.

At the end of the workshop, you should be able to:

  • understand what's needed to gather and analyze sensor data
  • create simple graphs and dashboards from sensor data
  • know the services SURF offers around IoT and sensor data

Prerequisites

We will build on the previous workshop on LoRa.

10:15 - 17:00 | Track 5. Artificial Intelligence and Machine Learning

10:15 - 11:30

Introduction and use cases of Machine Learning and Deep Learning

First presentation

Youssef El Bouhassani (Urban analytics HvA):
Introduction in Deep Learning 

 

Second presentation

Juan Rojo (Assistent Professor VU, Nikhef):
(Machine) Learning how to discover new particles at the LHC

A central component of the Large Hadron Collider (LHC) science program is the development of robust statistical tools that ensure an unbiased comparison between theoretical predictions and experimental data. Here I will discuss two applications of Machine Learning.  

At the end of this presentation you will understand how Machine Learning tools are used in applications at the high-energy frontier.
 

Third presentation

Faruk Diblen (Escience Research Engineer, Netherlands eScience Center)

DeePT: Deep learning for Proton Therapy - How AI can help for a better cancer treatment.

At the end of this talk you will show how AI can be useful to solve very complicated scientific problems and encourage researchers to use it to answer their scientific questions.

 

12:30 - 14:30

Machine Learning - part 1

Caspar van Leeuwen, Damian Podareanu - SURF Cluster Computing

In recent years machine learning and deep learning techniques in particular have developed tremendously. Neural networks are being used in more and more application domains going from computer vision to speech recognition, and even replacing parts of the compute pipeline for scientific HPC applications.

Learn how to use infrastructures efficiently to get the best performance out of different machine learning tools with several hands-on sessions.

15:00 - 17:00

Machine Learning - part 2

Continuation of part 1

11:30 - 12:30 | Lunch

11:30 - 12:30

Visit VR & AR demos

14:30 - 15:00 | Coffee / tea break

14:30 - 15:00

Visit VR & AR demos

17:00 - 18:00 | Drinks & snacks

17:00 - 18:00

Visit VR & AR demos

Latest modifications 29 Mar 2018