BigData@UGent in practice - session details

Yvan Saeys

Prof. dr. Yvan Saeys is group leader of the Datamining and Modeling for Biomedicine unit at the at the Inflammation Research Centre of VIB.

http://www.dambi.ugent.be

A gentle introduction to Big Data

The broad field of Big Data deals with data whose volume, diversity and complexity requires new techniques, algorithms and analyses to extract valuable knowledge from it. Big Data is currently revolutionizing many fields, both in science as well as industry, and in this talk I will provide a gentle introduction to the topic, particularly aiming at non-experts.

Download presentation slides

Jan Fostier

Prof. Dr. Ir. Jan Fostier is assistant professor at the IBCN research group at the department of Information Technology at Ghent University - iMinds.

https://www.ibcn.intec.ugent.be

Introduction to MapReduce and its Application to Post-Sequencing Analysis

This presentation serves as a gentle introduction to MapReduce, a programming model and software framework that can be used to process large datasets using parallel computing. Both the programming model and associated technical aspects such as the distributed filesystem are discussed in an entry-level manner. As an example showcase, I'll demonstrate that MapReduce is able to reduce the runtime from 12 days (sequential run) to less than three hours (parallel run) for a problem that is important for the bioinformatics community.

Download presentation slides

Ewald Pauwels

Dr. Ewald Pauwels is team leader of the High Performance Computing team (HPC-UGent) at the central Direction of Information and Communication Technology (DICT).

http://www.ugent.be/hpc

Big Data on the central UGent supercomputing infrastructure

In this presentation, I will briefly introduce the HPC-UGent team, its mission statement and role within the Flemish Supercomputer Center (VSC). I will specifically address the software and hardware supercomputing services that we offer to researchers with respect to Big Data projects. Finally, I will outline our future plans and investments.

Download presentation slides

Dirk Van den Poel

Prof. Dr. Dirk Van den Poel is Professor of Business Analytics/Big Data at Ghent University

www.bigdata.UGent.be

@dirkvandenpoel

dirk.vandenpoel@UGent.be

Spark: The New Kid on the Block

We start off with an overview of Spark (RDDs: Resilient Distributed Datasets) and the Spark Ecosystem, more specifically Spark SQL (Structured Query Language), MLLib (machine learning library). Next, the core Spark APIs will be discussed more in-depth. Some of the concepts will be illustrated in a real-life Big Data Analytics case study for a European retailer using IPython Notebooks.

Download presentation slides

Hiep Luong

Dr. Hiep Luong (postdoc in Wilfried Philips' research group, Department of Telecommunications and Information Processing)

From 15 MPixel images to 10 GPixel image data: big data in image
and video processing

As hardware is becoming better, cheaper and faster, the problems related to 'big data' have entered the domain of 'phenomenological image processing' as well as it did with other domains. Within the image and interpretation research group of UGent, we are confronted with requests to enhance, process and analyze such large datasets using only a limited amount of resources. In this presentation I will focus on the overall challenges we encounter with big image and video data. Finally, I will present Quasar, our in-house developed programming language that enables fast and efficient (parallel) computing on off-the-shelf hardware.

Download presentation slides

Frederik Coppens

Dr. Frederik Coppens

Data integration in life sciences: big data applications in plant research?

Due to the hype around Big Data, it is difficult to determine how these methodologies can be added value in life sciences. Here I want to give a overview of the thought process we went through at PSB before we decided to try out hadoop. I will also provide some examples where the paradigms used in Big Data could be applied in our research field.

Download presentation slides

Veronique Hoste

Prof. Dr. Veronique Hoste is head of the LT3 Language and Translation Technology Team at the faculty of Arts and Philosophy at Ghent University

How can Language Technology benefit from a supercomputing infrastructure?

In this talk, I will discuss the current trends and challenges in language technology and how a supercomputing infrastructure can support us in addressing these challenges.

Download presentation slides