Getting started with High-Performance Computing 2018-2019

Cluster

Research & Valorization

Target Group

Members of the Doctoral Schools of Ghent University

  • Part 1 of this course is intended for researchers that are new to working in a Unix/Linux environment. It comprises three full days of hands-on sessions where newly acquired skills will immediately be brought into practice. Participants must be entitled to request an account for the HPC infrastructure of the Vlaams Supercomputer Center (VSC). The number of participants for these sessions is restricted to 50.
  • Part 2 of this course is intended for researchers that are new to programming or new to Python programming. It comprises three full days of hands-on sessions where newly acquired skills will immediately be brought into practice through a series of case studies. The number of participants for these sessions is restricted to 50.

Level

All PhD students

Content of the course

Computers have played an increasingly important role in science for 50 years, and in particular the past decade and a half, and will continue to do so. Scientists today need to be completely computationally literate, as it simply becomes almost impossible to do competitive science without such literacy. This course provides an introduction the Unix command line, scripting and high-performance computing (HPC) as a stepping stone towards the use of computer clusters for solving advanced computation problems.

Programme

Part 1. Unix command line + shell scripting + HPC basics - 27-28-29 May 2019 (Mon-Tue-Wed = 3 days) - LUNCH IS INCLUDED

  • working interactively with the shell command line
  • consulting technical information
  • interactive text editing
  • file management
  • files and filters
  • protections and privileges
  • IO-redirection en pipes
  • regular expressions
  • process and job management
  • computer networks, X Window System
  • version control systems (http://github.ugent.be)
  • shell scripting basics

     - shell variables
     - processing command line arguments
     - control structures
     - file descriptors
     - shell functions

  • HPC Basics

     - submitting jobs
     - submitting array jobs
     - monitoring and managing submitted jobs
     - HPC environment variables
     - checkpointing


Part 2. Python scripting - 3-4-5 June 2019 ((Mon-Tue-Wed = 3 days) - LUNCH IS INCLUDED

  • introduction to Python
  • scientic computing with pylab (numpy, scipy & matplotlib) and Jupyter Notebook

Venue

The course will be organized in the multimedia room of the Information and Communication Technology Department (DICT) at Ghent University, as this facility has laptop computers for participants that cannot bring their own. Other course material will be made available through the electronic learning environment of Ghent University (http://zephyr.ugent.be)

Registration fee

Free of charge for Doctoral School members. The no show policy applies: no-show policy UGent

Registration

Please read the cancellation policy: cancellationpolicycourses

Teaching and learning material

  • Each participant of Part 1 will receive a personal copy of the textbook A Practical Guide to Linux Commands, Editors, and Shell Programming (3rd edition) by Mark G. Sobell, Prentice Hall (2013), ISBN-13: 978-0133085044
  • Each participant of Part 2 will receive a personal copy of the textbook Learning Python: Powerful Object-Oriented Programming (4th edition) by Mark Lutz, O'Reilly Media (2009), ISBN-13: 978-0596158064

References

E. Nemeth, G. Snyder, T. R. Hein, B. Whaley, UNIX and Linux System Administration Handbook (4th Edition), Prentice Hall, 2010.
A. Afzal, UNIX Unbounded: A Beginning Approach, fourth edition, Pearson Education, 2003.
D. Barron, The World of Scripting Languages, John Wiley & Sons, 2000.
R. Blum. Linux Command Line and Shell Scripting Bible, Wiley, 2008.
S.M. Sarwar, R. Koretsky, S.A. Sarwar, UNIX the textbook, second edition, Pearson Education, 2005.

Language

English

Evaluation methods and criteria (doctoral training programme)

Part I and Part II can be taken separately as 1 transferable skills seminar (cluster Research & Valorization) in the doctoral training programme after 100% participation