134.217 VSC-School Seminar for the
This course is in all assigned curricula part of the STEOP.
This course is in at least 1 assigned curriculum part of the STEOP.

2018W, SE, 2.0h, 3.0EC, to be held in blocked form

Properties

  • Semester hours: 2.0
  • Credits: 3.0
  • Type: SE Seminar

Aim of course

Improve the proficiency in High Performance Computing (HPC).

Subject of course


1st Block:            Linux and First Steps on the VSC-3 Cluster

Date and Time:     October 11, 2018, 14:00 - 18:00
Location:               TU Wien, FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)
Lecturers:              VSC Team (Claudia Blaas-Schenner, Siegfried Höfinger, Dieter Kvasnicka,
                              Balazs Lengyel, Irene Reichl, Siegfried Reinwald, Markus Stöhr, Jan Zabloudil)

Details and Registration see: http://vsc.ac.at/training/2018/VSC-Linux-Oct

Abstract:
This Linux command-line course is for users (or soon to be users) of the VSC-3 cluster only. You will learn how to login to VSC-3 and step-by-step we will show you how to work on the Linux command line and a few basic things that will help you to organise your workflows on the cluster. Focusing on hands-on teaching throughout the course, you will immediately try out what you've heard and adapt it to your own needs. After attending this course you are prepared for and might consider to continue with "Introduction to Working on the VSC-3 Cluster" to be able to use the VSC-3 supercomputer and especially its queuing system efficiently.


2nd Block:           Introduction to Working on the VSC-3 Cluster

Date and Time:     October 17, 2018, 09:00 - 16:00
Location:               TU Wien, FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)
Lecturers:              VSC Team (Claudia Blaas-Schenner, Siegfried Höfinger, Dieter Kvasnicka,
                              Balazs Lengyel, Irene Reichl, Siegfried Reinwald, Markus Stöhr, Jan Zabloudil)

Details and Registration see: http://vsc.ac.at/training/2018/VSC-Intro-Oct

Abstract:
In this course we will help you getting started on the VSC-3 cluster, Austria's most powerful supercomputer. With running and developing software on a supercomputer there are many similarities and fewer but crucial differences compared to your desktop PC. Focusing on hands-on teaching throughout the course, you will immediately try out what you've heard and adapt it to your own needs. This lecture is equally relevant to those who will merely be running existing software as to those who will develop scientific codes.


3rd Block:            Shared memory parallelization with OpenMP

Date and Time:      November 5 -6, 2018, 09:00 - 16:40
Location:               TU Wien, FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)
Lecturers:              Lukas Einkemmer  (lectures+practicals; Department of Mathematics, University of Innsbruck),
                              Claudia Blaas-Schenner and Irene Reichl (practicals only; VSC Team,  TU Wien)

Details and Registration see:  http://vsc.ac.at/training/2018/OpenMP-Nov

Abstract:
The focus of this 2 days course is on shared memory parallelization with OpenMP for dual-core, multi-core, shared memory, and ccNUMA platforms. This course teaches OpenMP starting from a beginners level. Hands-on sessions (in C and Fortran) will allow users to immediately test and understand the OpenMP directives, environment variables, and library routines. Race-condition debugging tools are also presented.


4th Block:            Parallelization with MPI

Date and Time:      November 7 - 9, 2018, 09:00 - 16:30 (9.11. until 16:00)
Location:               TU Wien, FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)
Lecturers:              Claudia Blaas-Schenner and Irene Reichl (VSC Team, TU Wien)

Details and Registration see: http://vsc.ac.at/training/2018/MPI-Nov

Abstract:
On clusters and distributed memory architectures, parallel programming with the Message Passing Interface (MPI) is the dominating programming model. This 3 days course teaches parallel programming with MPI starting from a beginners level. Hands-on sessions (in C and Fortran) will allow users to immediately test and understand the basic constructs of the Message Passing Interface (MPI).

5th Block:             Parallel I/O

Date and Time:      November 23, 2018 , 08:45 - 17:00
Location:               TU Wien, FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)
Lecturers:              Sebastian Lührs (Jülich Supercomputing Centre)

Details and Registration see: http://vsc.ac.at/training/2018/IO

Abstract:
This course will be about parallel I/O with a special focus on portable data formats. It will introduce the use of the HDF5 and NetCDF (NetCDF4 and PnetCDF) library interfaces, and hands-on exercises (in C/C++ or Fortran) will allow to immediately test and understand their usage. Performance hints, optimization potential, and best practices for I/O will be discussed in detail throughout the whole course.


6th Block:            Node-Level Performance Engineering

Date and Time:     December 5-7, 2018, 09:00 - 17:00
Location:               TU Wien, FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)
Lecturers:              Georg Hager and Gerhard Wellein (RRZE / HPC, Uni. Erlangen)

Details and Registration see: http://vsc.ac.at/training/2018/NLPE

Abstract:
This course covers performance engineering approaches on the compute node level. "Performance engineering" as we define it is more than employing tools to identify hotspots and bottlenecks. It is about developing a thorough understanding of the interactions between software and hardware. This process must start at the core, socket, and node level, where the code gets executed that does the actual computational work. Once the architectural requirements of a code are understood and correlated with performance measurements, the potential benefit of optimizations can often be predicted. We introduce a "holistic" node-level performance engineering strategy based on the Roofline model and apply it to different algorithms from computational science. Architectural details that are relevant for performance, such as pipelining, SIMD, superscalarity, memory hierarchies, etc., are covered in due detail.
This course provides –via lectures, demos, and hands-on labs– scientific training in Computational Science, and in addition, the scientific exchange of the participants among themselves.


7th Block:            Introduction to Working on the VSC-3 Cluster

Date and Time:     January 16, 2019, 09:00 - 16:00
Location:               TU Wien, FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)
Lecturers:              VSC Team (Claudia Blaas-Schenner, Siegfried Höfinger, Dieter Kvasnicka,
                              Balazs Lengyel, Irene Reichl, Siegfried Reinwald, Markus Stöhr, Jan Zabloudil)

Details and Registration see:  http://vsc.ac.at/training/2019/VSC-Intro-Jan

Abstract:
In this course we will help you getting started on the VSC-3 cluster, Austria's most powerful supercomputer. With running and developing software on a supercomputer there are many similarities and fewer but crucial differences compared to your desktop PC. Focusing on hands-on teaching throughout the course, you will immediately try out what you've heard and adapt it to your own needs. This lecture is equally relevant to those who will merely be running existing software as to those who will develop scientific codes.

 

Further topics/events will be announced as soon as they are scheduled.

Additional information

The VSC-School Seminar will be held in blocks, all interested scientists and students are welcome, course language is English.

Participation in selected single blocks is possible, please register for each block separately.

Contact for this course: vsc-seminar@list.tuwien.ac.at


Please consider the plagiarism guidelines of TU Wien when writing your seminar paper: http://www.tuwien.ac.at/fileadmin/t/ukanzlei/t-ukanzlei-english/Plagiarism.pdf


Please consider the plagiarism guidelines of TU Wien when writing your seminar paper: https://www.tuwien.ac.at/fileadmin/t/ukanzlei/t-ukanzlei-english/Plagiarism.pdf

Lecturers

Institute

Course dates

DayTimeDateLocationDescription
Thu14:00 - 18:0011.10.2018 FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Linux and First Steps on the VSC-3 Cluster
Wed09:00 - 16:0017.10.2018 FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Introduction to Working on the VSC-3 Cluster
09:00 - 16:4005.11.2018 - 06.11.2018 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Shared Memory Parallelization with OpenMP
09:00 - 16:3007.11.2018 - 09.11.2018 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Parallelization with MPI
Fri08:45 - 17:0023.11.2018 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Parallel I/O
09:00 - 17:0005.12.2018 - 07.12.2018 Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Node-Level Performance Engineering
Wed09:00 - 16:0016.01.2019 FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Introduction to Working on the VSC-3 Cluster
VSC-School Seminar for the - Single appointments
DayDateTimeLocationDescription
Thu11.10.201814:00 - 18:00 FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Linux and First Steps on the VSC-3 Cluster
Wed17.10.201809:00 - 16:00 FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Introduction to Working on the VSC-3 Cluster
Mon05.11.201809:00 - 16:40 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Shared Memory Parallelization with OpenMP
Tue06.11.201809:00 - 16:40 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Shared Memory Parallelization with OpenMP
Wed07.11.201809:00 - 16:30 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Parallelization with MPI
Thu08.11.201809:00 - 16:30 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Parallelization with MPI
Fri09.11.201809:00 - 16:30 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Parallelization with MPI
Fri23.11.201808:45 - 17:00 FH Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Parallel I/O
Wed05.12.201809:00 - 17:00 Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Node-Level Performance Engineering
Thu06.12.201809:00 - 17:00 Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Node-Level Performance Engineering
Fri07.12.201809:00 - 17:00 Internet-Raum FH1 (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Node-Level Performance Engineering
Wed16.01.201909:00 - 16:00 FH Schulungsraum ZID (TU Wien, Wiedner Hauptstraße 8-10, ground floor, red area)Introduction to Working on the VSC-3 Cluster
Course is held blocked

Examination modalities

Students (TU Wien and other Austrian universities) who want to earn credits (3 ECTS), please select a least 3 days according to your interest and send an email to: vsc-seminar@list.tuwien.ac.at

Course registration

Not necessary

Curricula

Literature

No lecture notes are available.

Previous knowledge

Prerequisites are good programming skills and some experience with running moderately large parallel codes.

Language

English