• PATCs and PTCs

  • PRACE operates six PRACE Advanced Training Centres (PATCs) since 2012, and they have established a state-of-the-art curriculum for training in HPC and scientific computing. PATCs carry out and coordinate training and education activities that enable both European academic researchers and European industry to utilise the computational infrastructure available through PRACE and provide top-class education and training opportunities for computational scientists in Europe.

    The six PRACE Advanced Training Centres (PATCs) are based at:

    • Barcelona Supercomputing Center (Spain)
    • Consorzio Interuniversitario, CINECA (Italy)
    • CSC – IT Center for Science Ltd (Finland)
    • EPCC at the University of Edinburgh (UK)
    • Gauss Centre for Supercomputing (Germany)
    • Maison de la Simulation (France)

    In addition to operating the PATCs, 4 PRACE Training Centres (PTCs) will be piloted. The PTCs will expand the geographical reach of PATCs by sourcing PATC courses locally, collaborating with PATCs in delivering courses locally or by complementing the PATC programme with local courses.

    The four selected PRACE Training Centers (PTCs) are based at:

    • GRNET – Greek Research and Technology Network (Greece)
    • ICHEC – Irish Centre for High-End Computing (Ireland)
    • IT4I – National Supercomputing Center VSB Technical University of Ostrava (Czech Republic)
    • SURFsara (The Netherlands)

    The following figure depicts the location of the PATC and PTC centers throughout Europe.

    PATC events this month:

    December 2017
    Mon Tue Wed Thu Fri Sat Sun
     
    The registration to this course is now open. Please, bring your own laptop. All the PATC courses at BSC are free of charge.

    Course conveners: Andrea Manrique and Louis-Philippe Caron

    Course lecturers: Francesco Benincasa, Miguel Castrillo, Neven Fučkar, Oriol Jorba Casellas, Joan López de la Franca, Domingo Manubens, Nicolau Manubens, Kim Serradell Maronda, Oriol Tinto

    Objectives: The objective of this PATC course is to cover the basics of a high performance computing (HPC) environment oriented towards earth science applications, specifically chemical weather modelling and climate modelling.

    More precisely, the course will cover:

    Introduction to earth science fundamentals and modelling;
    Basic usage of an HPC environment: shell, compilers, libraries, file systems, queuing system;
    Build and configure targeted earth science applications with the NMMB/BSC-CTM chemical transport model and with the EC-EARTH climate model;
    Execute and monitor numerical experiments using a workflow manager;
    Analyse and visualise model outputs with a wide set of tools.
    Learning outcomes: Participants will learn and gain experience in accessing an HPC facility, installing earth science numerical models and related utilities and libraries, running numerical simulations, monitoring the execution of supercomputing jobs, analysing and visualising model results.

    Level: (All courses are designed for specialists with at least 1st cycle degree or similar background experience)
    For trainees with some theoretical and practical knowledge

    Prerequisites:
    At least University degree in progress on Earth Sciences, Computer Sciences or related area
    Basic knowledge of UNIX
    Knowledge of C, FORTRAN, MPI or openMP is recommended
    Knowledge of Earth Sciences data formats is recommended (grib, netcdf, hdf,…)

    Agenda:

    Day 1 (Nov. 29)
    10:00-11:30 Introduction to earth science fundamentals and modelling.    11:30-11:50 Coffee break
    11:50-13:30 Introduction to Linux and HPC, BSC facilities, PRACE and RES    13:30-14:30 Lunch break
    14:30-16:00 HPC hands-on tutorial. At the end of this tutorial, the students will be able
    to compile and execute jobs on Mare Nostrum 4.    16:00-16:20 Coffee break
    16:20-18:00 Introduction to analysis and visualisation tools for model outputs
    (s2dverification, MapGenerator, CDO, NCO, panoply, ncview)    20:00 Social dinner (participants pay individually)

    Day 2 (Nov 30)
    10:00-11:00 Lecture on the NMMB-MONARCH chemical transport model.
    11:00-12:00 Lecture on setting up and running the model.    12:00-12:30 Coffee break
    12:30-14:00 Model hands-on tutorial. Students will run a prepared case.    14:00-15:00 Lunch break
    15:00-17:00 Analysis hands-on tutorial. Students will apply tools for analysis and
    visualisation to the outputs created in the morning.

    Day 3 (Dec 1)
    10:00-11:00 Lecture on the EC-EARTH climate model.
    11:00-12:00 Lecture on setting up and running the model with Autosubmit.
        12:00-12:30 Coffee break
    12:30-14:00 Model hands-on tutorial. Students will run with a prepared case.    14:00-15:00 Lunch break
    15:00-17:00 Analysis hands-on tutorial. Students will apply tools for analysis and
    visualisation to the outputs created in the morning.
    17:00-17:30 MareNostrum visit

    End of Course

    https://events.prace-ri.eu/event/642/
    Nov 29 9:00 to Dec 1 18:00
    This course provides an introduction to High-Performance Computing (HPC) for researchers in the life sciences, using ARCHER as a platform for hands-on training exercises.

    The course is organised and funded by BioExcel - the Centre of Excellence for Computational Biomolecular Research (bioexcel.eu) and PRACE, and delivered in collaboration with ARCHER - the UK national supercomputing service (archer.ac.uk).

    Overview

    High-performance computing (HPC) is a fundamental technology used to solve a wide range of scientific research problems. Many important challenges in science such as protein folding, the search for the Higgs boson, drug discovery, and the development of nuclear fusion all depend on simulations, models and analyses run on HPC facilities to make progress.
     
    This course introduces HPC to life science researchers, focusing on the aspects that are most important for those new to this technology to understand. It will help you judge how HPC can best benefit your research, and equip you to go on to successfully and efficiently make use of HPC facilities in future. The course will cover basic concepts in HPC hardware, software, user environments, filesystems, and programming models. It also provides an opportunity to gain hands-on practical experience and assistance using an HPC system (ARCHER, the UK national supercomputing service) through examples drawn from the life sciences, such as biomolecular simulation.

    Registration

    Registration is on a first-come, first-served basis via https://events.prace-ri.eu/event/626/registration/

    Learning outcomes

    On completion of the course, we expect that attendees will understand and be able to explain:
     
        • Why HPC? - What are the drivers and motivation? Who uses it and why?
        • The UK & EU HPC landscape - HPC facilities available to researchers
        • HPC hardware - Building blocks and architectures
        • Parallel computing - Programming models and implementations
        • Using HPC systems
            • Access
            • Batch schedulers & resource allocation
            • Running jobs
            • Dealing with errors
            • Compiling code
            • Using libraries
            • Performance
        • The Future of HPC

    Pre-requisites

    This course follows on naturally from the BioExcel Summer School on Foundation skills for HPC in computational biomolecular research (bioexcel.eu/events/bioexcel-summer-school/)

    Familiarity with basic Linux commands (at the level of being able to navigate a file system) is recommended. You may find a Linux 'cheat sheet' such as www.archer.ac.uk/docume.....ckref useful if you are less familiar with Linux.

    No programming skills or previous HPC experience is required.

    Desktop computers running Windows will be available, however you are encouraged to bring your own laptop (running Windows, Linux, or macOS) as you will find it useful to learn how to set this up to connect to ARCHER (with assistance from course helpers if needed) and perform the hands-on practicals.

    Bursaries

    BioExcel will be providing a limited number of fixed amount travel bursaries for this event. If you would like to be considered for a travel bursary email Vera Matser (matser [at] ebi.ac.uk). Eligibility criteria and conditions for the travel grants are available at bioexcel.eu/wp-content/.....t.pdf
     

    Timetable

    Wednesday November 29th

    10:00 - Welcome, introduction, and course overview
    Review of HPC skills and competencies framework
    Familiarisation with fellow attendees

    11:00 - LECTURE - What is HPC?
    11:25 - PRACTICAL - Connecting to ARCHER
    11:30 - BREAK - Coffee & Tea
    12:00 - PRACTICAL - Sequence Alignment
    13:00 - BREAK - Lunch
    14:00 - LECTURE - Parallel Computing Patterns
    14:30 - LECTURE - Measuring Parallel Performance
    15:00 - PRACTICAL - Sequence Alignment (continued)
    15:30 - BREAK - Coffee & Tea
    16:00 - PRACTICAL - Sequence Alignment (continued)
    16:15 - LECTURE - Building Blocks - Software (Operating System, Processes and Threads)
    16:45 - LECTURE - Building Blocks - Hardware (Processors & cores, Memory, Accelerators)
    17:15 - Review of the day
    17:30 - Finish

    Thursday November 30th

    9:30 - Summary of day 1
    9:45 - LECTURE - Parallel Models
    10:30 - PRACTICAL - Fractal
    11:00 - BREAK - Coffee & Tea
    11:30 - PRACTICAL - Fractal (continued)
    12:00 - LECTURE - HPC Architectures
    12:30 - LECTURE - Batch Systems & Parallel Application Launchers
    13:00 - BREAK - Lunch
    14:00 - PRACTICAL - Molecular Dynamics
    15:00 - LECTURE - Compilers and Building Software
    15:30 - BREAK - Coffee & Tea
    16:00 - PRACTICAL - Molecular Dynamics (continued)
    16:30 - LECTURE - Parallel libraries
    17:00 - Review of the day
    17:15 - Finish

    19:00 - Dinner at Amber restaurant

    Friday December 1st

    9:30 - Summary of day 2
    9:45 - LECTURE - Pipelines and workflows
    10:15 - PRACTICAL - Molecular Dynamics (continued)
    11:00 - LECTURE - The UK & EU HPC Landscape
    11:30 - BREAK - Coffee & Tea
    12:00 - LECTURE - The Future of HPC
    12:30 - LECTURE - "Where next?" and things to remember
    13:00 - Lunch
    14:00 - Individual consultations, course review and feedback / competency survey
    15:00 - Finish

    Course Materials

    www.archer.ac.uk/traini.....x.php

     

    Dinner

    The course dinner will be held at the Amber restaurant, located at the Scotch Whisky Experience near Edinburgh Castle, on Thursday 30th November at 19:00.
    For location and travel information see https://www.scotchwhiskyexperience.co.uk/plan-your-visit

     

     

     

    https://events.prace-ri.eu/event/626/
    Nov 29 11:00 to Dec 1 16:00
    This course teaches 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, apply it to different algorithms from computational science, and also show how an awareness of the performance features of an application may lead to notable reductions in power consumption.

    The course is a PRACE Advanced Training Center event.

    Introduction
    Our approach to performance engineering
    Basic architecture of multicore systems: threads, cores, caches, sockets, memory
    The important role of system topology

    Tools: topology & affinity in multicore environments
    Overview
    likwid-topology and likwid-pin

    Microbenchmarking for architectural exploration
    Properties of data paths in the memory hierarchy
    Bottlenecks
    OpenMP barrier overhead

    Roofline model: basics
    Model assumptions and construction
    Simple examples
    Limitations of the Roofline model

    Pattern-based performance engineering
    Optimal use of parallel resources
    Single Instruction Multiple Data (SIMD)
    Cache-coherent Non-Uniform Memory Architecture (ccNUMA)
    Simultaneous Multi-Threading (SMT)

    Tools: hardware performance counters
    Why hardware performance counters?
    likwid-perfctr
    Validating performance models

    Roofline case studies
    Dense matrix-vector multiplication
    Sparse matrix-vector multiplication
    Jacobi (stencil) smoother

    Optional: The ECM performance model


    https://events.prace-ri.eu/event/632/
    Nov 30 9:00 to Dec 1 17:00
    2
     
    3
     
    Description

    This course gives a thorough introduction to programming GPUs using the directive based OpenACC paradigm. The course consists of lectures and hands-on exercises. Topics of this course include the basic usage of OpenACC, as well as some more advanced issues related to profiling, performance and interoperability with CUDA.

    Learning outcome

    After the course the participants should have the basic skills needed for utilizing OpenACC in new, or existing programs.

    Prerequisites

    The participants are assumed to have working knowledge of Fortran and/or C programming languages. In addition, fluent operation in a Linux/Unix environment will be assumed.

    Agenda

    Day 1, Monday 4.12

    09:00 - 12:00 SESSION 1 & Coffee break

    Introduction to accelerators​
    Introduction to OpenACC
    Exercises
    12:00 - 13:00 Lunch

    13:00 - 16:00 SESSION 2 & Coffee break

    Data movement
    Exercises
    Day 2, Tuesday 5.12

    09:00 - 12:00 SESSION 3 & Coffee break

    Profilling
    Performance considerations
    Exercises
    12:00 - 13:00 Lunch

    13:00 - 16:00 SESSION 4 & Coffee break

    Asynchronous operations and pipelining
    Interoperability with CUDA and GPU-Accelerated libraries
    Exercises
     

    Lecturers:   Sami Ilvonen (CSC), Fredrik Robertsén (CSC) 

    Language:  EnglishPrice:          Free of charge

    https://events.prace-ri.eu/event/670/
    Dec 4 8:00 to Dec 5 15:00
    This training will present basic elements first to enable developpers to understand when and how to use a debugger, and second to optimize the performance of their codes.

    Debugging :

    A short introduction to debugging with examples
    Optimization :

     Basics (stripping, IOs, getting the calculations out of the loops, changing the algorithm, ...)
     Compiler options - Profilers (Valgrind, Intel Vtune)
     Vectorization - Data access (cache usage maximization)
     OpenMP parallelization strategies
     Theory to upper-bound the expected performance benefit (speedup, efficiency, peak, memory bandwidth, ...)
    Half of the course will be made of hands-on sessions.Learning outcomes 
    Ability to understand the main issues for code optimization, knowledge of the main tools and techniques for basic debugging.Prerequisites 
    Basic knowledge of Unix, programming experience in C or Fortran, OpenMp

    https://events.prace-ri.eu/event/664/
    Dec 4 14:00 to Dec 8 12:00
    This training will present basic elements first to enable developpers to understand when and how to use a debugger, and second to optimize the performance of their codes.

    Debugging :

    A short introduction to debugging with examples
    Optimization :

     Basics (stripping, IOs, getting the calculations out of the loops, changing the algorithm, ...)
     Compiler options - Profilers (Valgrind, Intel Vtune)
     Vectorization - Data access (cache usage maximization)
     OpenMP parallelization strategies
     Theory to upper-bound the expected performance benefit (speedup, efficiency, peak, memory bandwidth, ...)
    Half of the course will be made of hands-on sessions.Learning outcomes 
    Ability to understand the main issues for code optimization, knowledge of the main tools and techniques for basic debugging.Prerequisites 
    Basic knowledge of Unix, programming experience in C or Fortran, OpenMp

    https://events.prace-ri.eu/event/664/
    Dec 4 14:00 to Dec 8 12:00
    Description

    This course gives a thorough introduction to programming GPUs using the directive based OpenACC paradigm. The course consists of lectures and hands-on exercises. Topics of this course include the basic usage of OpenACC, as well as some more advanced issues related to profiling, performance and interoperability with CUDA.

    Learning outcome

    After the course the participants should have the basic skills needed for utilizing OpenACC in new, or existing programs.

    Prerequisites

    The participants are assumed to have working knowledge of Fortran and/or C programming languages. In addition, fluent operation in a Linux/Unix environment will be assumed.

    Agenda

    Day 1, Monday 4.12

    09:00 - 12:00 SESSION 1 & Coffee break

    Introduction to accelerators​
    Introduction to OpenACC
    Exercises
    12:00 - 13:00 Lunch

    13:00 - 16:00 SESSION 2 & Coffee break

    Data movement
    Exercises
    Day 2, Tuesday 5.12

    09:00 - 12:00 SESSION 3 & Coffee break

    Profilling
    Performance considerations
    Exercises
    12:00 - 13:00 Lunch

    13:00 - 16:00 SESSION 4 & Coffee break

    Asynchronous operations and pipelining
    Interoperability with CUDA and GPU-Accelerated libraries
    Exercises
     

    Lecturers:   Sami Ilvonen (CSC), Fredrik Robertsén (CSC) 

    Language:  EnglishPrice:          Free of charge

    https://events.prace-ri.eu/event/670/
    Dec 4 8:00 to Dec 5 15:00
    This training will present basic elements first to enable developpers to understand when and how to use a debugger, and second to optimize the performance of their codes.

    Debugging :

    A short introduction to debugging with examples
    Optimization :

     Basics (stripping, IOs, getting the calculations out of the loops, changing the algorithm, ...)
     Compiler options - Profilers (Valgrind, Intel Vtune)
     Vectorization - Data access (cache usage maximization)
     OpenMP parallelization strategies
     Theory to upper-bound the expected performance benefit (speedup, efficiency, peak, memory bandwidth, ...)
    Half of the course will be made of hands-on sessions.Learning outcomes 
    Ability to understand the main issues for code optimization, knowledge of the main tools and techniques for basic debugging.Prerequisites 
    Basic knowledge of Unix, programming experience in C or Fortran, OpenMp

    https://events.prace-ri.eu/event/664/
    Dec 4 14:00 to Dec 8 12:00
    This training will present basic elements first to enable developpers to understand when and how to use a debugger, and second to optimize the performance of their codes.

    Debugging :

    A short introduction to debugging with examples
    Optimization :

     Basics (stripping, IOs, getting the calculations out of the loops, changing the algorithm, ...)
     Compiler options - Profilers (Valgrind, Intel Vtune)
     Vectorization - Data access (cache usage maximization)
     OpenMP parallelization strategies
     Theory to upper-bound the expected performance benefit (speedup, efficiency, peak, memory bandwidth, ...)
    Half of the course will be made of hands-on sessions.Learning outcomes 
    Ability to understand the main issues for code optimization, knowledge of the main tools and techniques for basic debugging.Prerequisites 
    Basic knowledge of Unix, programming experience in C or Fortran, OpenMp

    https://events.prace-ri.eu/event/664/
    Dec 4 14:00 to Dec 8 12:00
    8
     
    9
     
    10
     
    The two centres of excellence in computer application EoCoE and POP will join forces once more to hold a hands-on workshop on HPC benchmarking and performance analysis from 11-14 December 2017. The fourth event in the series started two years ago, it will be held at Maison de la Simulation near Paris, this time supported by the French PATC. 

                                                      

     

    This particular workshop is especially designed to welcome engineers/researchers from the industry who are developing and/or using their own numerical tools. They will be able to apply a unique performance analysis methodology to their application to permit seamless monitoring of over 20 performance indicators. Tools and results from the workshop can be taken back home in order to benchmark future software modifications in a reproducible manner, allowing very significant productivity improvement during the development and optimization phase. Participants will also have ample time to meet and discuss with HPC experts from the EoCoE and POP projects to seek expertise on special issues or to eventually engage into further collaboration.

    The previous workshops have so far brought together 17 experts from topical fields in energy research and  tools and 11 experts from HPC science in order to tackle the transition of current R&D codes and applications towards exascale. The EoCoE performance analysis methodology, with its post-workshop 'aftercare' service, has reached a new level of maturity and recognition. Many applications have already achieved significant (2x-10x) speed-ups since being examined at one of these events.

    Registration is free but mandatory.

    Monday PM: JUBE + hands-on
    Tuesday AM: Score-P, Scalasca, Vampir + hands-on
    Tuesday PM: Extrae, Paraver + hands-on
    Wednesday AM: Tools integration hands-on
    Wednesday PM: Tools integration hands-on
    Thursday AM: Tools integration hands-on


    https://events.prace-ri.eu/event/666/
    Dec 11 to Dec 13
    The two centres of excellence in computer application EoCoE and POP will join forces once more to hold a hands-on workshop on HPC benchmarking and performance analysis from 11-14 December 2017. The fourth event in the series started two years ago, it will be held at Maison de la Simulation near Paris, this time supported by the French PATC. 

                                                      

     

    This particular workshop is especially designed to welcome engineers/researchers from the industry who are developing and/or using their own numerical tools. They will be able to apply a unique performance analysis methodology to their application to permit seamless monitoring of over 20 performance indicators. Tools and results from the workshop can be taken back home in order to benchmark future software modifications in a reproducible manner, allowing very significant productivity improvement during the development and optimization phase. Participants will also have ample time to meet and discuss with HPC experts from the EoCoE and POP projects to seek expertise on special issues or to eventually engage into further collaboration.

    The previous workshops have so far brought together 17 experts from topical fields in energy research and  tools and 11 experts from HPC science in order to tackle the transition of current R&D codes and applications towards exascale. The EoCoE performance analysis methodology, with its post-workshop 'aftercare' service, has reached a new level of maturity and recognition. Many applications have already achieved significant (2x-10x) speed-ups since being examined at one of these events.

    Registration is free but mandatory.

    Monday PM: JUBE + hands-on
    Tuesday AM: Score-P, Scalasca, Vampir + hands-on
    Tuesday PM: Extrae, Paraver + hands-on
    Wednesday AM: Tools integration hands-on
    Wednesday PM: Tools integration hands-on
    Thursday AM: Tools integration hands-on


    https://events.prace-ri.eu/event/666/
    Dec 11 to Dec 13
    OpenMP is the industry standard for shared-memory programming, which enables serial programs to be parallelised using compiler directives.This course is aimed at programmers seeking to deepen their understanding of OpenMP and explore some of its more recent and advanced features.

    This 3-day course will cover topics including nested parallelism, OpenMP tasks, the OpenMP memory model, performance tuning, hybrid OpenMP + MPI, OpenMP implementations, and new features in OpenMP 4.0/4.5. Hands-on practical programming exercises make up a significant, and integral, part of this course.

    Attendees should be familiar with the basics of OpenMP, including parallel regions, data scoping, work sharing directives and synchronisation constructs. Access will be given to appropriate hardware for all the exercises, although many of them can also be performed on a standard Linux laptop.

     

    Pre-course setup

    All attendees should bring their own wireless-enabled laptop. Practical exercises will be done using a guest account on ARCHER. You will need an ssh client such as terminal on a Mac or Linux machine, or putty or MobaXterm on Windows. The course tutor will be able to assist with settings to connect on the day. You should also have a web browser, a pdf reader and a simple text editor.

     

    Timetable

    Day 1

    09:00 - 11:00  Lectures: OpenMP basics: Parallel regions, Worksharing, Synchronisation
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practical: Parallel regions
    13:00 - 14:00  Lunch
    14:00 - 15:30  Lectures: Multicore and multithreaded CPUs, Caches, Cache coherency, NUMA
    15:30 - 16:00  Tea
    16:00 - 17:00  Practicals: Streams, Coherency

    Day 2

    09:00 - 11:00  Lectures: Tasks, Nested parallelism, Memory model
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practicals: Mandelbrot with nested loops, collapse, and tasks
    13:00 - 14:00  Lunch
    14:00 - 15:30  Lectures: OpenMP + MPI
    15:30 - 16:00  Tea
    16:00 - 17:00  Practicals: OpenMP + MPI

    Day 3

    09:00 - 11:00  Lectures: OpenMP tips, tricks and pitfalls, Performance issues
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practicals: MD tuning
    13:00 - 14:00  Lunch
    14:00 - 15:30  OpenMP 4.0/4.5 features, alternative APIs
    15:30 - 16:00  Tea
    16:00 - 17:00  Consultancy session: bring your own code!

    Course Materials

    www.archer.ac.uk/traini.....x.php

     

    https://events.prace-ri.eu/event/637/
    Dec 12 10:00 to Dec 14 18:30
    The two centres of excellence in computer application EoCoE and POP will join forces once more to hold a hands-on workshop on HPC benchmarking and performance analysis from 11-14 December 2017. The fourth event in the series started two years ago, it will be held at Maison de la Simulation near Paris, this time supported by the French PATC. 

                                                      

     

    This particular workshop is especially designed to welcome engineers/researchers from the industry who are developing and/or using their own numerical tools. They will be able to apply a unique performance analysis methodology to their application to permit seamless monitoring of over 20 performance indicators. Tools and results from the workshop can be taken back home in order to benchmark future software modifications in a reproducible manner, allowing very significant productivity improvement during the development and optimization phase. Participants will also have ample time to meet and discuss with HPC experts from the EoCoE and POP projects to seek expertise on special issues or to eventually engage into further collaboration.

    The previous workshops have so far brought together 17 experts from topical fields in energy research and  tools and 11 experts from HPC science in order to tackle the transition of current R&D codes and applications towards exascale. The EoCoE performance analysis methodology, with its post-workshop 'aftercare' service, has reached a new level of maturity and recognition. Many applications have already achieved significant (2x-10x) speed-ups since being examined at one of these events.

    Registration is free but mandatory.

    Monday PM: JUBE + hands-on
    Tuesday AM: Score-P, Scalasca, Vampir + hands-on
    Tuesday PM: Extrae, Paraver + hands-on
    Wednesday AM: Tools integration hands-on
    Wednesday PM: Tools integration hands-on
    Thursday AM: Tools integration hands-on


    https://events.prace-ri.eu/event/666/
    Dec 11 to Dec 13
    OpenMP is the industry standard for shared-memory programming, which enables serial programs to be parallelised using compiler directives.This course is aimed at programmers seeking to deepen their understanding of OpenMP and explore some of its more recent and advanced features.

    This 3-day course will cover topics including nested parallelism, OpenMP tasks, the OpenMP memory model, performance tuning, hybrid OpenMP + MPI, OpenMP implementations, and new features in OpenMP 4.0/4.5. Hands-on practical programming exercises make up a significant, and integral, part of this course.

    Attendees should be familiar with the basics of OpenMP, including parallel regions, data scoping, work sharing directives and synchronisation constructs. Access will be given to appropriate hardware for all the exercises, although many of them can also be performed on a standard Linux laptop.

     

    Pre-course setup

    All attendees should bring their own wireless-enabled laptop. Practical exercises will be done using a guest account on ARCHER. You will need an ssh client such as terminal on a Mac or Linux machine, or putty or MobaXterm on Windows. The course tutor will be able to assist with settings to connect on the day. You should also have a web browser, a pdf reader and a simple text editor.

     

    Timetable

    Day 1

    09:00 - 11:00  Lectures: OpenMP basics: Parallel regions, Worksharing, Synchronisation
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practical: Parallel regions
    13:00 - 14:00  Lunch
    14:00 - 15:30  Lectures: Multicore and multithreaded CPUs, Caches, Cache coherency, NUMA
    15:30 - 16:00  Tea
    16:00 - 17:00  Practicals: Streams, Coherency

    Day 2

    09:00 - 11:00  Lectures: Tasks, Nested parallelism, Memory model
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practicals: Mandelbrot with nested loops, collapse, and tasks
    13:00 - 14:00  Lunch
    14:00 - 15:30  Lectures: OpenMP + MPI
    15:30 - 16:00  Tea
    16:00 - 17:00  Practicals: OpenMP + MPI

    Day 3

    09:00 - 11:00  Lectures: OpenMP tips, tricks and pitfalls, Performance issues
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practicals: MD tuning
    13:00 - 14:00  Lunch
    14:00 - 15:30  OpenMP 4.0/4.5 features, alternative APIs
    15:30 - 16:00  Tea
    16:00 - 17:00  Consultancy session: bring your own code!

    Course Materials

    www.archer.ac.uk/traini.....x.php

     

    https://events.prace-ri.eu/event/637/
    Dec 12 10:00 to Dec 14 18:30
    Description:

    This course focuses on the development and execution of bioinformatics pipelines and on their optimization with regards to computing time and disk space. In an era where the data produced per-analysis is in the order of terabytes, simple serial bioinformatic pipelines are no longer feasible. Hence the need for scalable, high-performance parallelization and analysis tools which can easily cope with large-scale datasets. To this end we will study the common performance bottlenecks emerging from everyday bioinformatic pipelines and see how to strike down the execution times for effective data analysis on current and future supercomputers.

    As a case study, two different bioinformatics pipelines (whole-exome and transcriptome analysis) will be presented and re-implemented on the supercomputers of Cineca thanks to ad-hoc hands-on sessions aimed at applying the concepts explained in the course.

    Topics: 

    NGS-data, big-data analysis, code parallelization, MPI, running a bioinformatics pipeline, large-scale sample datasets.

    Target audience: 

    Biologists, bioinformaticians and computer scientists interested in approaching large-scale NGS-data analysis for the first time.

    Pre-requisites: 

    Basic knowledge of python and command line. A very basic knowledge of biology is recommended but not required.

    Grant
    The lunch for the three days will be offered to all the participants and some grants are available. The only requirement to be eligible is to be not funded by your institution to attend the course and to work or live in an institute outside the Roma area. The grant  will be 300 euros for students working and living outside Italy and 150 euros for students working and living in Italy (except Rome area). Some documentation will be required and the grant will be paid only after a certified presence of minimum 80% of the lectures.

    Further information about how to request the grant, will be provided at the confirmation of the course: about 3 weeks before the starting date.

    https://events.prace-ri.eu/event/661/
    Dec 13 9:00 to Dec 15 17:00
    OpenMP is the industry standard for shared-memory programming, which enables serial programs to be parallelised using compiler directives.This course is aimed at programmers seeking to deepen their understanding of OpenMP and explore some of its more recent and advanced features.

    This 3-day course will cover topics including nested parallelism, OpenMP tasks, the OpenMP memory model, performance tuning, hybrid OpenMP + MPI, OpenMP implementations, and new features in OpenMP 4.0/4.5. Hands-on practical programming exercises make up a significant, and integral, part of this course.

    Attendees should be familiar with the basics of OpenMP, including parallel regions, data scoping, work sharing directives and synchronisation constructs. Access will be given to appropriate hardware for all the exercises, although many of them can also be performed on a standard Linux laptop.

     

    Pre-course setup

    All attendees should bring their own wireless-enabled laptop. Practical exercises will be done using a guest account on ARCHER. You will need an ssh client such as terminal on a Mac or Linux machine, or putty or MobaXterm on Windows. The course tutor will be able to assist with settings to connect on the day. You should also have a web browser, a pdf reader and a simple text editor.

     

    Timetable

    Day 1

    09:00 - 11:00  Lectures: OpenMP basics: Parallel regions, Worksharing, Synchronisation
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practical: Parallel regions
    13:00 - 14:00  Lunch
    14:00 - 15:30  Lectures: Multicore and multithreaded CPUs, Caches, Cache coherency, NUMA
    15:30 - 16:00  Tea
    16:00 - 17:00  Practicals: Streams, Coherency

    Day 2

    09:00 - 11:00  Lectures: Tasks, Nested parallelism, Memory model
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practicals: Mandelbrot with nested loops, collapse, and tasks
    13:00 - 14:00  Lunch
    14:00 - 15:30  Lectures: OpenMP + MPI
    15:30 - 16:00  Tea
    16:00 - 17:00  Practicals: OpenMP + MPI

    Day 3

    09:00 - 11:00  Lectures: OpenMP tips, tricks and pitfalls, Performance issues
    11:00 - 11:30  Coffee
    11:30 - 13:00  Practicals: MD tuning
    13:00 - 14:00  Lunch
    14:00 - 15:30  OpenMP 4.0/4.5 features, alternative APIs
    15:30 - 16:00  Tea
    16:00 - 17:00  Consultancy session: bring your own code!

    Course Materials

    www.archer.ac.uk/traini.....x.php

     

    https://events.prace-ri.eu/event/637/
    Dec 12 10:00 to Dec 14 18:30
    Description:

    This course focuses on the development and execution of bioinformatics pipelines and on their optimization with regards to computing time and disk space. In an era where the data produced per-analysis is in the order of terabytes, simple serial bioinformatic pipelines are no longer feasible. Hence the need for scalable, high-performance parallelization and analysis tools which can easily cope with large-scale datasets. To this end we will study the common performance bottlenecks emerging from everyday bioinformatic pipelines and see how to strike down the execution times for effective data analysis on current and future supercomputers.

    As a case study, two different bioinformatics pipelines (whole-exome and transcriptome analysis) will be presented and re-implemented on the supercomputers of Cineca thanks to ad-hoc hands-on sessions aimed at applying the concepts explained in the course.

    Topics: 

    NGS-data, big-data analysis, code parallelization, MPI, running a bioinformatics pipeline, large-scale sample datasets.

    Target audience: 

    Biologists, bioinformaticians and computer scientists interested in approaching large-scale NGS-data analysis for the first time.

    Pre-requisites: 

    Basic knowledge of python and command line. A very basic knowledge of biology is recommended but not required.

    Grant
    The lunch for the three days will be offered to all the participants and some grants are available. The only requirement to be eligible is to be not funded by your institution to attend the course and to work or live in an institute outside the Roma area. The grant  will be 300 euros for students working and living outside Italy and 150 euros for students working and living in Italy (except Rome area). Some documentation will be required and the grant will be paid only after a certified presence of minimum 80% of the lectures.

    Further information about how to request the grant, will be provided at the confirmation of the course: about 3 weeks before the starting date.

    https://events.prace-ri.eu/event/661/
    Dec 13 9:00 to Dec 15 17:00
    Description:

    This course focuses on the development and execution of bioinformatics pipelines and on their optimization with regards to computing time and disk space. In an era where the data produced per-analysis is in the order of terabytes, simxle sarIal bioinformatic pipelines are no longer feasible. Hence the need for scalable, high-performance parallelization and analysis tools which can easily cope with large-scale datasets. To this end we will study the common performance bottlenecks emerging from everyday bioinformatic pipelines and see how to strike down the execution times for effective data analysis on current and future supercomputers.

    As a case study, two different bioinformatics pipelines (whole-exome and transcriptome analysis) will be presented and re-implemented on the supercomputers of Cineca thanks to ad-hoc hands-on sessions aimed at applying the concepts explained in the course.

    Topics: 

    NGS-data, big-data analysis, code parallelization, MPI, running a bioinformatics pipeline, large-scale sample datasets.

    Target audience: 

    Biologists, bioinformaticians and computer scientists interested in approaching large-scale NGS-data analysis for the first time.

    Pre-requisites: 

    Basic knowledge of python and command line. A very basic knowledge of biology is recommended but not required.

    Grant
    The lunch for the three days will be offered to all the participants and some grants are available. The only requirement to be eligible is to be not funded by your institution to attend the course and to work or live in an institute outside the Roma area. The grant  will be 300 euros for students working and living outside Italy and 150 euros for students working and living in Italy (except Rome area). Some documentation will be required and the grant will be paid only after a certified presence of minimum 80% of the lectures.

    Further information about how to request the grant, will be provided at the confirmation of the course: about 3 weeks before the starting date.

    https://events.prace-ri.eu/event/661/
    Dec 13 9:00 to Dec 15 17:00
    16
     
    17
     
    18
     
    19
     
    20
     
    21
     
    22
     
    23
     
    24
     
    25
     
    26
     
    27
     
    28
     
    29
     
    30
     
    31
     


    PTC events this month:

    December 2017
    Mon Tue Wed Thu Fri Sat Sun
     
    Introduction to Parallel Programming

    29 November  - 1 December 2017

    Purpose of the course

    The focus is to understand the basics of parallel programming with the message-passing interface (MPI) and OpenMP parallel programming paradigms. MPI is the dominant parallelization paradigm in high performance computing and enables one to write programs that run on distributed memory machines, such as the ARIS Greek supercomputer and other systems of the PRACE infrastructure. OpenMP is a threading based approach which enables one to parallelize a program over a single shared memory machine, such as a single node in ARIS. The course also contains performance and best practice considerations, e.g., hybrid MPI+OpenMP parallelisation. The course ends with a section presenting profiling and code optimisations to understand the behavior and performance of parallelized codes.

    The 3 day course consists of lectures and hands-on exercises on parallel programming. Hands-on sessions (in C and Fortran) will allow users to immediately test and understand the taught constructs of the Message Passing  Interface (MPI) and the shared memory directives of OpenMP. The course ends with a section presenting profiling and code optimizations to understand the behavior and performance of parallelized codes.

    After the course the participants should be able to write simple parallel programs and parallelize existing programs with the basic features of MPI or OpenMP.

    Prerequisites

    The course addresses participants who are familiar with the C/C++ or Fortran programming languages and have working experience with the Linux operating system and the use of the command line. Experience with parallel programming (MPI and OpenMP) is not required.

    Bring your own laptop in order to be able to participate in the training hands on. Hands on work will be done in pairs so if you don’t have a laptop you might work with a colleague.

    Course language is English.

    Registration

    The maximum number of participants is 25.

    Registrations will be evaluated on a first-come, first-served basis. GRNET is responsible for the selection of the participants on the basis of the training requirements and the technical skills of the candidates. GRNET will also seek to guarantee the maximum possible geographical coverage with the participation of candidates from many countries.

    Venue

    GRNET headquarters

    Address: 2nd  Floor, 7, Kifisias Av. GR 115 23 Athens

    Information on how to reach GRNET headquarters ia available on GRNET website: https://grnet.gr/en/contact-us/  

    Accommodation options near GRNET can be found at: https://grnet.gr/wp-content/uploads/sites/13/2015/11/Hotels-near-GRNET-en.pdf

    ARIS - System Information

    ARIS is the name of the Greek supercomputer, deployed and operated by GRNET (Greek Research and Technology Network) in Athens. ARIS consists of 532 computational nodes seperated in four “islands” as listed here:


    426 thin nodes: Regular compute nodes without accelerator.


    44 gpu nodes: “2 x NVIDIA Tesla k40m” accelerated nodes.


    18 phi nodes: “2 x INTEL Xeon Phi 7120p” accelerated nodes.


    44 fat nodes: Fat compute nodes have larger number of cores and memory per core than a thin node.

    All the nodes are connected via Infiniband network and share 2PB GPFS storage.The infrastructure also has an IBM TS3500 library of maximum storage capacity of about 6 PB. Access to the system is provided by two login nodes.

    About Tutors

    Dr. Dellis (Male) holds a B.Sc. in Chemistry (1990) and PhD in Computational Chemistry (1995) from the National and Kapodistrian University of Athens, Greece. He has extensive HPC and grid computing experience. He was using HPC systems in computational chemistry research projects on fz-juelich machines (2003-2005). He received an HPC-Europa grant on BSC (2009). In EGEE/EGI projects he acted as application support and VO software manager for SEE VO, grid sites administrator (HG-02, GR-06), NGI_GRNET support staff (2008-2014). In PRACE 1IP/2IP/3IP/4IP/5IP he was involved in benchmarking tasks either as group member or as BCO (2010-2017). Currently he holds the position of “Senior HPC Applications Support Engineer” at GRNET S.A. where he is responsible for activities related to user consultations, porting, optimization and running HPC applications at national and international resources.

    Dr. Ioannis E. Venetis received his PhD in 2006 from the Computer Engineering and Informatics Department at the University of Patras, Greece. Currently he teaches "Parallel Processing" and "Software and Programming for High Performance Systems" at the same Department. He has participated in numerous research projects in the area of Parallel Computing. His main research interests include parallel programming models, run-time systems for supporting such models, co-processor programming (especially using GPUs and the Intel Xeon Phi) and parallelization of computationally demanding applications.

    Paschalis Korosoglou is the technical coordinator of the Scientific Computing Office at the Aristotle University of Thessaloniki. He holds an MSc in Computational science and has participated in major national and European projects, such as EGEE, EGI, SEE-GRID, EUChinaGrid, HellasGrid, PRACE and HellasHPC, mainly in user services provisioning and application support activities. In this respect he holds a solid background on scientific applications porting, profiling, optimisation and benchmarking activities and has developed under several occasions code optimisation improvements related to hybrid parallel implementations (MPI/OpenMP) and parallel I/O best practices. He has been involved in national and international Working Groups, such as CSIRT, EUGridPMA, GOC, SAM and is an active contributor within the e-Infrastructures Policy Group (e-IRG).

    George Z. Zachos is a student at the Department of Computer Science & Engineering, University of Ioannina. He has extensive experience on systems administration and systems programming. Since 2015 he is a member of the Parallel Processing Group of the same Department where he works on shared and distributed memory systems and their programming models. Currently, he is actively involved with the development of OMPi, an open source OpenMP compiler and runtime system for C.

    About GRNET

    GRNET provides Internet connectivity, high-quality e-Infrastructures and advanced services to the Greek Educational, Academic and Research community.

    Through its high-speed, high-capacity infrastructure that spans across the entire country, GRNET interconnects more than 150 institutions, including all universities and technological institutions, as well as many research institutes and the public Greek School Network.

    GRNET operates the National High Performance Computing system (a Tier-1 in the European HPC ecosystem) and offers user and application support services, that provide Greek scientists with the computing infrastructure and expertise they need for their research enabling them to perform large scale simulations.

    GRNET offers innovative IaaS cloud computing services to the Greek and global research & education communities: “ ~okeanos” and “okeanos global” allow users to create multi-layer virtual infrastructure and instantiate virtual computing machines, local networks to interconnect them, and a reliable storage space within seconds, with few, simple mouse clicks.

    GRNET aims at contributing towards Greece’s Digital Convergence with the EU, by supporting the development and encouraging the use of e-Infrastructures and services. The right and timely planning strategies, together with the long experience and know-how of its people, guarantee the continuation and enhancement of GRNET’s successful course.

    Greek Research and Technology Network – Networking Reserach and Education:

    www.grnet.gr, hpc.grnet.gr

    https://events.prace-ri.eu/event/673/
    Nov 29 10:00 to Dec 1 15:00
    2
     
    3
     
    4
     
    5
     
    6
     
    7
     
    8
     
    9
     
    10
     
    11
     
    12
     
    13
     
    Summary

    The course provides an introduction to the OpenFOAM open source package for CFD simulations on HPC infrastructure. The participants will be guided through basic geometry and mesh generation, physical modelling, visualization, and parallel computation. In particular, the agenda includes the following training blocks:

    introduction
    geometry and meshing
    mesh manipulations
    physical modelling and numerical simulations
    parallel computing
    post-processing
    Purpose of the course (benefits for the attendees)

    The course provides a tour around all fundamental parts of OpenFOAM. It is useful for new users and beginners in CFD simulations as it uses open source code.

    About the tutor

    Tomáš Brzobohatý is a researcher at the Czech IT4Innovations National Supercomputing Center. He is a member a team which provides services for industrial partners in the field of CFD and CSM analysis.

    https://events.prace-ri.eu/event/636/
    Dec 14 9:00 to Dec 15 15:30
    Summary

    The course provides an introduction to the OpenFOAM open source package for CFD simulations on HPC infrastructure. The participants will be guided through basic geometry and mesh generation, physical modelling, visualization, and parallel computation. In particular, the agenda includes the following training blocks:

    introduction
    geometry and meshing
    mesh manipulations
    physical modelling and numerical simulations
    parallel computing
    post-processing
    Purpose of the course (benefits for the attendees)

    The course provides a tour around all fundamental parts of OpenFOAM. It is useful for new users and beginners in CFD simulations as it uses open source code.

    About the tutor

    Tomáš Brzobohatý is a researcher at the Czech IT4Innovations National Supercomputing Center. He is a member a team which provides services for industrial partners in the field of CFD and CSM analysis.

    https://events.prace-ri.eu/event/636/
    Dec 14 9:00 to Dec 15 15:30
    16
     
    17
     
    18
     
    19
     
    20
     
    21
     
    22
     
    23
     
    24
     
    25
     
    26
     
    27
     
    28
     
    29
     
    30
     
    31