PRACE

Partnership for Advanced Computing in Europe

PRACE https://tess.elixir-europe.org/content_providers/prace Partnership for Advanced Computing in Europe /system/content_providers/images/000/000/059/original/1c2d412b-c20c-43b8-926d-6febf042f341?1479828059
Found 0 materials.
Showing 14 upcoming events. Found 835 past events. View all results.
  • [ONLINE] Directive-based GPU programming with OpenACC @ JSC

    27 - 29 October 2021

    [ONLINE] Directive-based GPU programming with OpenACC @ JSC https://tess.elixir-europe.org/events/online-directive-based-gpu-programming-with-openacc-jsc-cc1f4c22-ba41-4772-9cc5-ced2b4f659b7 GPU-accelerated computing drives current scientific research. Writing fast numeric algorithms for GPUs offers high application performance by offloading compute-intensive portions of the code to the GPU. The course will cover basic aspects of GPU architectures and programming. Focus is on the usage of the directive-based OpenACC programming model which allows for portable application development. Examples of increasing complexity will be used to demonstrate optimization and tuning of scientific applications. Topics covered will include: Introduction to GPU/Parallel computing Programming model OpenACC Interoperability of OpenACC with GPU libraries (like cuBLAS and cuFFT) and CUDA Multi-GPU Programming with MPI and OpenACC Tools for debugging and profiling Performance optimization The course consists of lectures and interactive hands-on sessions in C or Fortran (the attendee’s choice). This course will take place as an online event. The link to the online platform will be provided to the registrants only. Prerequisites: Some knowledge about Linux, e.g. make, command line editor, Linux shell (see for instance this overview), a little experience in C/Fortran Application Registrations are only considered until 1 October  2021, the maximal number of participants is limited. Applicants will be notified three weeks before the course, whether they are accepted for participation. The link to the course platform will be sent shortly before the course. Instructors: Dr. Andreas Herten, Dr. Thorsten Hater, Dr. Kaveh Haghighi-Mood, JSC; Markus Hrywniak, Jiri Kraus, NVIDIA Contact For any questions concerning the course please send an e-mail to a.herten@fz-juelich.de https://events.prace-ri.eu/event/1253/ 2021-10-27 07:00:00 UTC 2021-10-29 11:00:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] SYCL Workshop @ENCCS

    8 - 9 November 2021

    [ONLINE] SYCL Workshop @ENCCS https://tess.elixir-europe.org/events/sycl-workshop Overview Welcome to this ENCCS workshop, in collaboration with CSC Finland and Izum, Slovenia. The training will run on the EuroHPC JU Vega system. SYCL_ is a C++ abstraction layer for programming heterogeneous hardware with a single-source approach. SYCL_ is high-level, cross-platform, and extends standard ISO C++17. You will learn to: Use the hipSYCL_ compiler to generate executable for multiple hardware targets. Write hardware-agnostic code to express parallelism using the queue, command group, and kernel abstractions. Use buffer and accessors to handle memory across devices. Evaluate drawbacks and advantages of unified shared memory. Agenda For updated agenda, please visit https://enccs.se/events/2021/11/sycl-workshop/ https://events.prace-ri.eu/event/1257/ 2021-11-08 08:00:00 UTC 2021-11-09 11:10:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] Advanced Deep Learning with Transformers @ENCCS&RISE

    21 October 2021

    [ONLINE] Advanced Deep Learning with Transformers @ENCCS&RISE https://tess.elixir-europe.org/events/online-advanced-deep-learning-with-transformers-enccs-rise Overview In recent years, Graph Neural Networks (GNNs) and Transformers have led to numerous break-through achievements in a variety of fields such as Natural Language Processing (NLP), chemistry and physics. By doing away with the need for fixed-size inputs, these architectures significantly extend the scope of problems in which deep learning can be applied. Preliminary Agenda This workshop will take you from the representation of graphs and finite sets as inputs for neural networks to the implementation of full GNNs for a variety of tasks. You will learn about the central concepts used in GNNs in a hands-on setting using Jupyter Notebooks and a series of coding exercises. While the workshop will use problems from the field of chemistry as an example for applications, the skills you learn can be transferred to any domain where finite set or graph-based representations of data are appropriate. From GNNs, we will make the leap to Transformer architectures, and explain the conceptual ties between the two. The workshop is free of charge and will be conducted fully online using zoom. Prerequisites To successfully participate in this workshop, you should have a good understanding of basic linear algebra and core concepts of deep learning such as CNNs, stochastic gradient descent, and supervised learning. You should also be familiar with the implementation of neural networks using PyTorch. A basic conceptual understanding of mathematical graphs is recommended but not a prerequisite. Agenda For updated agenda you can follow this link https://enccs.se/events/2021/10/advanced-deep-learning/ Contact person Apostolos Vasileiadis apostolos.vasileiadis@it.uu.se https://events.prace-ri.eu/event/1260/ 2021-10-21 07:00:00 UTC 2021-10-21 10:30:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] Data Science and Deep Learning with Python @ Cineca

    22 - 26 November 2021

    [ONLINE] Data Science and Deep Learning with Python @ Cineca https://tess.elixir-europe.org/events/online-data-science-and-deep-learning-with-python-cineca Description: The increasing amount of data collected through sensors or computational simulations can take advantage of new techniques for being processed in order to extract new insights out of raw data. The purpose of this course is to present researchers with data science methods and techniques. The course includes theoretical lectures followed by practical sessions on data manipulation, visualisation, machine learning and deep learning.  A specific session on how to use HPC resources is included. Participants will utilize Python on their personal computers as well as Cineca HPC resources for practical projects. Topics: -Introduction to data analytics -Machine Learning theory (supervised/unsupervised) -Data Manipulation with Numpy and Pandas -Data visualization with Matplotlib, Seaborn, Bokeh -Machine learning with sci-kit learn -Deep learning with Tensorflow and Keras -Introduction to Cineca HPC facilities Target Audience: Students and researchers from academia and industry with different backgrounds who are looking for technologies and methods to analyse large amounts of data. Prerequisites: Participants must have basic knowledge in statistics, fundamentals of computer programming with Python and use of GNU/Linux-based systems. The number of participants is limited to 25 students. Applicants will be selected according to their experience, qualification and scientific interest BASED ON WHAT WRITTEN IN THE "Reason for participation" FIELD OF THE REGISTRATION FORM.  APPLICATION DEADLINE OCT 29th, 2021.  STUDENTS WILL BE NOTIFIED ON THEIR ADMISSION OR NOT WITH AN EMAIL ON MONDAY NOV 9th.  Attendance is FREE.    https://events.prace-ri.eu/event/1261/ 2021-11-22 08:00:00 UTC 2021-11-26 17:00:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] High Performance Bioinformatics@CINECA

    1 - 3 December 2021

    [ONLINE] High Performance Bioinformatics@CINECA https://tess.elixir-europe.org/events/online-high-performance-bioinformatics-cineca-cc74092e-a7db-4dbd-a99f-f769dc572523 We live in a big-data era and simple serial bioinformatic pipelines can’t efficiently handle huge input datasets. Hence, High Performance Computing (HPC) can represent a good solution for researchers who need to analyze and address new biological questions with their data.  This course is both theoretical and practical and is addressed to bioinformaticians who want to scale up their analysis on a cluster machine. It mainly focuses on the development and execution of automated data analysis pipelines. On the first day, students will become confident with a cluster machine (e.g. hardware, software, module environment, data storage) and will learn how to submit a single batch scripts via the SLURM scheduler. On the second day, partecipants will be introduced to the world of Next Generation Sequencing (NGS) and will learn how to build a fully automated RNA-seq pipeline able to handle large input datasets, focusing on job concatenations and HPC request resource optimization. On the last day students will be introduced to cloud computing and to the world of snakemake, a python workflow management system tool able to create reproducible and scalable data analyses, with particular attention to scaling workflows on cluster and cloud without modifyng the workflow definition. Ad-hoc hands-on sessions, aimed at applying the concepts explained during the course, will be held every afternoon. Skills: By the end of the course each student should be able to: - Know all the conventions and opportunities offered by CINECA for accessing HPC resources; - Download datasets from public repositories and/or transfer input files from the user’s local computer to the CINECA clusters; - Navigate through the software environment set up by CINECA; - Run single-step jobs on a supercomputer via SLURM scheduler; - Combine several bioinformatics applications into a fully automated pipeline able to run on a supercomputer; - Learn how to iterate through samples in order to manage huge input datasets; - Have an overview of how to take advantage of snakemake to build a portable, scalable and fully automated pipeline. Target audience: Biologists, bioinformaticians and computer scientists interested in approaching large-scale NGS-data analysis. Course prerequisites: Good knowledge of python and shell command line. A very basic knowledge of R and biology is recommended but not strictly required. https://events.prace-ri.eu/event/1262/ 2021-12-01 08:00:00 UTC 2021-12-03 15:00:00 UTC [] [] [] workshops_and_courses [] []
  • Programming Distributed Computing Platforms with COMPSs @ BSC

    25 - 26 January 2022

    Programming Distributed Computing Platforms with COMPSs @ BSC https://tess.elixir-europe.org/events/programming-distributed-computing-platforms-with-compss-bsc-c0947e02-6579-44e3-b688-8abacab9d005 Please, bring your own laptop. All the PATC courses at BSC are free of charge. Course convener: Rosa Badia, Workflows and Distributed Computing Group Manager, Computer Sciences - Workflows and Distributed Computing Department Lecturers:  Rosa M Badia, Workflows and Distributed Computing Group Manager, Computer Sciences - Workflows and Distributed Computing Department, BSC Javier Conejero, Senior Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC Jorge Ejarque, Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC Daniele Lezzi, Senior Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC Francesc Lordan, Postdoctoral Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC Objectives: The objective of this course is to give an overview of the COMPSs programming model, which is able to exploit the inherent concurrency of sequential applications and execute them in a transparent manner to the application developer in distributed computing platform. This is achieved by annotating part of the code as tasks, and building at execution a task-dependence graph based on the actual data used consumed/produced by the tasks. The COMPSs runtime is able to schedule the tasks in the computing nodes and take into account facts like data locality and the different nature of the computing nodes in case of heterogeneous platforms. Additionally, recently COMPSs has been enhanced with the possibility of coordinating Web Services as part of the applications. COMPSs supports Java, C/C++ and Python as programming languages. Learning Outcomes:  In the course, the COMPSs syntax, programming methodology and an overview of the runtime internals will be given. The attendees will get a first lesson about programming with COMPSs that will enable them to start programming with this framework. A hands-on with simple introductory exercises will be also performed. The students who finish this course will be able to develop simple COMPSs applications and to run them both in a local resource and in a distributed platform (initially in a private cloud). The exercises will be delivered in Python and Java. In case of Python, Jupyter notebooks will be used in some of the exercises. Level: for trainees with some theoretical and practical knowledge. INTERMEDIATE: for trainees with some theoretical and practical knowledge; those who finished the beginners course ADVANCED: for trainees able to work independently and requiring guidance for solving complex problems Prerequisites: Programming skills in Java and Python  Tentative Agenda: Day 1 (January 25th, 2022) 9:30 - 10:00 - Welcome and round table (30') 10:00 - 10:30 - Session 1: Introduction to COMPSs (30') 10:30 - 11:15 - Session 2: PyCOMPSs: Writing Python applications (45') 11:15 - 11:45 - Coffee break 11:45 - 13:00 - Session 3: Python Hands-on using Jupyter notebooks (1h15') 13:00 - 14:30 - Lunch break 14:30 - 15:00 - Session 4: Machine learning with dislib (30') 15:00 - 16:30 - Session 5: Hands-on with dislib (1h30') Day 2 (January 26th, 2022) 9:30 - 11:00 - Session 6: Java & C++ (1h30') 11:00 - 11:30 - Cofee break 11:30 - 13:00 - Session 7: COMPSs Advanced Features (1h30') 13:00 - 14:30 - Lunch break 14:30 - 16:30 - Session 8: Cluster Hands-on (MareNostrum) (2h) 16:30 - 16:45 - Session 9: COMPSs Installation & Final Notes (15') END of COURSE https://events.prace-ri.eu/event/1233/ 2022-01-25 08:30:00 UTC 2022-01-26 15:30:00 UTC [] [] [] workshops_and_courses [] []
  • Managing distributed data with Hecuba and dataClay @ BSC

    27 January 2022

    Managing distributed data with Hecuba and dataClay @ BSC https://tess.elixir-europe.org/events/managing-distributed-data-with-hecuba-and-dataclay-bsc-b47a99b3-4cfe-4b3b-bb06-b529efc290ff Please, bring your own laptop. All the PATC courses at BSC are free of charge. Course conveners: Department and Research group: Computer Science - Workflows and Distributed Computing Yolanda Becerra, Data-driven Scientific Computing research line, Senior researcher Anna Queralt, Distributed Object Management research line, Senior researcher Course Lecturers: Department and Research group: Computer Sciences - Workflows and Distributed Computing Alex Barceló, Distributed object Management research line, Researcher Yolanda Becerra, Data-driven Scientific Computing research line, Senior researcher Daniel Gasull, Distributed object Management research line, Research engineer Juanjo Costa, Computer Architecture Department at UPC Anna Queralt, Distributed object Management research line, Senior researcher Objectives: The objective of this course is to give an overview of BSC storage solutions, Hecuba and dataClay. These two platforms allow to easily store and manipulate distributed data from object-oriented applications, enabling programmers to handle object persistence using the same classes they use in their programs, thus avoiding time consuming transformations between persistent and non-persistent data models. Also, Hecuba and dataClay enable programmers to transparently manage distributed data, without worrying about its location. This is achieved by adding a minimal set of annotations in the classes. Both Hecuba and dataClay can work independently or integrated with the COMPSs programming model and runtime to facilitate parallelization of applications that handle persistent data, thus providing a comprehensive mechanism that enables the efficient usage of persistent storage solutions from distributed programming environments. Both platforms offer a common interface to the application developer that facilitates using one solution or the other depending on the needs, without changing the application code. Also, both of them have additional features that allow the programmer to take advantage of their particularities. Learning Outcomes:   In the course, the Hecuba and dataClay syntax, programming methodology and an overview of their internals will be given. Also, an overview of COMPSs at user level will be provided in order to take advantage of the distribution of data with both platforms. The attendees will get a first lesson about programming with the common storage interface that will enable them to start programming with both frameworks. A hands-on with simple introductory exercises will be also performed for each platform, with and without COMPSs to distribute the computation. The students who finish this course will be able to develop simple Hecuba and dataClay applications and to run them both in a local resource and in a distributed platform (initially in a private cloud) Prerequisites: Basic programming skills in Python and Java. Previous attendance to PATC course on programming distributed systems with COMPSs is recommended. Tentative Agenda:   Day 1 (Jan 27) Session 1 / 9:30 – 13:00 9:30-10:00 Round table. Presentation and background of participants 10:00-11:00 Motivation, introduction and syntax of BSC storage platforms 11:00-11:30 Coffee break 11:30-12:15 Hands-on with storage API 12:15-13:00 COMPSs overview and how to parallelize a sequential application 13:00-14:30 Lunch break Session 2/ 14:30 – 18:00 14:30-16:00 Hecuba specifics and hands-on 16:00-16:30 Break 16:30-18:00 dataClay specifics and hands-on       https://events.prace-ri.eu/event/1234/ 2022-01-27 09:00:00 UTC 2022-01-27 17:00:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] Earth Sciences Simulation Environments @ BSC

    9 - 11 November 2021

    [ONLINE] Earth Sciences Simulation Environments @ BSC https://tess.elixir-europe.org/events/online-earth-sciences-simulation-environments-bsc The registration to this course will open soon. Please, bring your own laptop. All the PATC courses at BSC are free of charge. Course conveners: Herve Petetin and Vladimir Lapin This training will be online via zoom. Course lecturers: Earth Sciences Department researchers involved in the group of Computational Earth Sciences, Atmospheric Composition, Climate Prediction and Earth System Services. Final lectures TBA. 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 and parallel computing. 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,…) Knowledge of R and Python Agenda: Agenda: Day 1 (Nov. 09) – 4.5h 10:00-10:15 Welcome and introduction of the BSC Earth Sciences Department. (Francisco Doblas-Reyes, Head of BSC-ES Department) 10:15-10:30 Technical and logistic information. (Hervé Petetin and Vladimir Lapin) 10:30-11:00 Introduction to earth science fundamentals and modelling. (Eleftheria Exarchou) 11:00-11:30 Break 11:30-13:00 Introduction to Linux and HPC, BSC facilities, PRACE and RES + preparation/checks for hands-on tutorial after lunch break (Mario Acosta) 13:00-14:00 Lunch break 14:00-16:00 HPC hands-on tutorial. At the end of this tutorial, the students will be able to compile and execute jobs on Mare Nostrum 4. (Mario Acosta, Miguel Castrillo and Hervé Petetin and Vladimir Lapin) Day 2 (Nov 10) – 3.5h 9:30-10:00 Lecture on the EC-EARTH climate model. (Eleftheria Exarchou) 10:00-10:30 Lecture on the NMMB-MONARCH atmosphere chemistry model. (Enza di Tomaso and Jan Mateu) 10:30-11:00 Lecture on setting up and running the NMMB-MONARCH model with Autosubmit. (Francesca Macchia and Gilbert Montane) 11:00-11:30 Break 11:30-13:30 Model hands-on tutorial. Students will run a prepared case. (Francesca Macchia, Gilbert Montane, Hervé Petetin and Vladimir Lapin) Day 3 (Nov 11) – 3h 10:00-11:00 Introduction to analysis and visualisation tools for model outputs (s2dverification, MapGenerator, CDO, NCO, panoply, ncview). (Francesco Benincasa) 11:00-11:30 Break 11:30-13:30 Analysis hands-on tutorial. Students will apply tools for analysis and visualisation to the outputs created the day before. (Amalia Vradi, An-Chi Ho, Nuria Perez, Hervé Petetin and Vladimir Lapin)   END of COURSE https://events.prace-ri.eu/event/1231/ 2021-11-09 09:00:00 UTC 2021-11-11 12:30:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] Advanced Fortran Topics @ LRZ

    23 - 26 November 2021

    [ONLINE] Advanced Fortran Topics @ LRZ https://tess.elixir-europe.org/events/online-advanced-fortran-topics-lrz-a524a65e-54e4-45a8-a557-56e13b64dc0e Overview This online course is targeted at scientists who wish to extend their knowledge of Fortran to cover advanced features of the language. Topics covered include (still preliminary) Best Practices global objects and interfaces abstract interfaces and the IMPORT statement object based programming Object-Oriented Programming type extension, polymorphism and inheritance binding of procedures to types and objects generic type-bound procedures abstract types and deferred bindings IEEE features and floating point exceptions Interoperability with C mixed language programming patterns Fortran 2003 I/O extensions OO Design Patterns: application of object-oriented programming creation and destruction of objects polymorphic objects and function arguments interacting objects dependency inversion: submodules and plugins Coarrays PGAS concepts and coarray basics dynamic entities advanced synchronization parallel programming patterns recent enhancements: collectives, events, teams, atomic subroutines performance aspects of coarray programming The course combines lectures and hands-on sessions. Prerequisites Course participants should have good knowledge of the Fortran 95 standard (such as covered in the February course at LRZ), and also basic UNIX/Linux knowledge (e.g. shell commands, simple scripts, editors like vi or emacs). Hands-On Please use your own laptop or institute cluster for the hands-on sessions. On your laptop or PC you should run either a Linux distribution, or a Microsoft Linux for Windows subsystem. A recent version of the gfortran compiler should be installed. Alternatively, an evaluation version of the NAG compiler (version 7.0) can be used. The latter can be obtained from https://www.nag.com/content/software-trials?product=NAG%20Compiler. Finally, the GSL (GNU scientific library) and its header files should be installed. Language English Lecturers Dr. R. Bader, Dr. G. Brietzke (LRZ) Prices and Eligibility The course is open and free of charge for people from academia and industry. Registration Via https://events.prace-ri.eu/event/1244/registrations/913/ Please register with your official e-mail address to prove your affiliation. Withdrawal Policy See Withdrawal https://events.prace-ri.eu/event/1244/ 2021-11-23 09:00:00 UTC 2021-11-26 15:00:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] PATC: Introduction into the Big Data Analytics @ BSC

    7 - 11 February 2022

    [ONLINE] PATC: Introduction into the Big Data Analytics @ BSC https://tess.elixir-europe.org/events/online-patc-introduction-into-the-big-data-analytics-bsc All the PATC courses at BSC are free of charge. Course Convener:  Maria-Ribera Sancho Course Lecturers: To be determined Objectives: The course brings together key information technologies used in manipulating, storing, and analysing data including: the basic tools for statistical analysis techniques for parallel processing tools for access to unstructured data storage solutions Learning outcomes: Students will be introduced to systems that can accept, store, and analyse large volumes of unstructured data. The learned skills can be used in data intensive application areas. Level: For trainees with some theoretical and practical knowledge   Agenda: To be determined https://events.prace-ri.eu/event/1245/ 2022-02-07 08:30:00 UTC 2022-02-11 15:30:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] ChEESE Advanced Training on HPC for Computational Seismology @ BSC

    23 - 25 November 2021

    [ONLINE] ChEESE Advanced Training on HPC for Computational Seismology @ BSC https://tess.elixir-europe.org/events/parallel-programming-workshop-bsc-c0c4f034-83ff-4b55-b1bd-caab3b02034b The registration to this course will open soon. Please, bring your own laptop. All the PATC courses at BSC are free of charge. Course Convener: To be determined Location: Online via Zoom Objectives: Are you interested in numerical wave simulations but feel like there are too many codes and methods out there and don’t know which option is right for you? Or perhaps you just want to refresh your knowledge and gain some practical experience with the different possibilities? Either way, this fully virtual workshop is for you! This training course will present introductory hands-on tutorials for ExaHyPE, Salvus, SeisSol, and SPECFEM3D, which are four different high-performance computational seismology software packages focused on waveform modeling and inversion. The course will provide you with an overview of their similarities, differences, and areas of applicability. Unique and in-depth half-day practicals will be provided for each package, with computational resources provided by BSC. Participants will also be encouraged to present their own work and/or motivation as a one-minute lightning-talk during the wrap-up session of the workshop.   This training event is hosted by ChEESE, the Center of Excellence for Exascale in Solid Earth.   Requirements: - Basic knowledge of computational seismology (e.g., familiarity with the elastic wave equation) - Basic knowledge of Python and Jupyter notebooks - Optional: basic programming skills (C/C++, only for the ExaHyPE course module) Learning Outcomes: - A basic introduction to waveform physics modeling - Hands-on experience with four different wave propagation codes Agenda: To be determined   https://events.prace-ri.eu/event/904/ 2021-11-23 08:30:00 UTC 2021-11-25 16:30:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] Node-Level Performance Engineering @ LRZ

    1 - 3 December 2021

    [ONLINE] Node-Level Performance Engineering @ LRZ https://tess.elixir-europe.org/events/online-node-level-performance-engineering-lrz-5b04463a-1fa4-44e4-92ad-c37fd2f4b0a0 This online course organised in cooperation with NHR@FAU covers performance engineering approaches on the compute node level. Even application developers who are fluent in OpenMP and MPI often lack a good grasp of how much performance could at best be achieved by their code. This is because parallelism takes us only half the way to good performance. Even worse, slow serial code tends to scale very well, hiding the fact that resources are wasted. This course conveys the required knowledge to develop 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. We introduce the basic architectural features and bottlenecks of modern processors and compute nodes. Pipelining, SIMD, superscalarity, caches, memory interfaces, ccNUMA, etc., are covered. A cornerstone of node-level performance analysis is the Roofline model, which is introduced in due detail and applied to various examples from computational science. We also show how simple software tools can be used to acquire knowledge about the system, run code in a reproducible way, and validate hypotheses about resource consumption. Finally, once the architectural requirements of a code are understood and correlated with performance measurements, the potential benefit of code changes can often be predicted, replacing hope-for-the-best optimizations by a scientific process. The course is a PRACE training event. Introduction Basic architecture of multicore systems: pipelines, SIMD, caches, sockets, memory The important role of system topology Tools: topology & affinity in multicore environments likwid-topology and likwid-pin, alternatives Roofline model basics          Model assumptions and construction Simple examples Limitations of the Roofline model Tools: hardware performance counters Why hardware performance counters? likwid-perfctr Validating performance models Optimal use of parallel resources Single Instruction Multiple Data (SIMD) Cache-coherent Non-Uniform Memory Architecture (ccNUMA) Roofline case studies Tall & skinny dense matrix-matrix multiplication Sparse matrix-vector multiplication Jacobi (stencil) smoother Basics of performance engineering PE process Code profiling Proper benchmarking Reproducibility and documentation Beyond Roofline: The ECM performance model (optional) Prerequisites You have to be able to handle a Linux command line and file editing remotely. Basic knowledge of C, C++, or Fortran programming and of OpenMP is required. Hands-On Exercises will be done on a cluster at NHR@FAU. For the exercises you need an SSH client on your local computer. Language English Lecturers Prof. Gerhard Wellein, Dr. habil. Georg Hager (RRZE, Uni. Erlangen) Prices and Eligibility The course is open and free of charge for people from academia and industry. Withdrawal Policy See Withdrawal Legal Notices This course is offered in cooperation with NHR@FAU. Some of your personal data will be transferred to NHR@FAU (title, first name, surname, institution, country, email, course) to create the course accounts. The legal basis is in accordance with Article 6(1)(b) GDPR. Please see also our data protection notice (in German: https://www.lrz.de/datenschutzerklaerung/). https://events.prace-ri.eu/event/1249/ 2021-12-01 08:00:00 UTC 2021-12-03 15:00:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] Shared memory parallelization with OpenMP @ VSC Vienna

    28 - 29 October 2021

    [ONLINE] Shared memory parallelization with OpenMP @ VSC Vienna https://tess.elixir-europe.org/events/online-shared-memory-parallelization-with-openmp-vsc-vienna-6a1f95a7-ee21-4079-9033-8d906aed3059 This course will be delivered as an ONLINE COURSE (using Zoom) for remote participation. REGISTRATION is strictly NECESSARY since the details to access the online course will be provided to the registered and accepted attendees only. Registration follows the "first come – first served" principle. For the online course we can take a maximum of 40 persons. Additional registrations will be added to a waiting list. Furthermore, accepted participants will be contacted a few days before the course and asked to do a short PRE-ASSIGNMENT that has to be completed before the course starts. For doing the HANDS-ON LABS all you need is a C/C++ or Fortran compiler that supports OpenMP. Participants can choose between using their own laptops or any system they have access to that provides an OpenMP enabled compiler, it's even possible to do the exercises in Google Colab. Alternatively, participants can ask for a training user account at the VSC for the duration of the course. Overview 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. Content Levels: Beginners = 6:00h (50%) + Intermediate = 4:00h (33%) + Advanced = 2:00h (17%) This course is a PRACE training event. It is organized by the VSC Research Center, TU Wien, in cooperation with the High-Performance Computing-Center Stuttgart (HLRS). Agenda & Content 1st day 08:45  Join online 09:00    Welcome 09:10    Introduction to modern hardware (talk) 10:10    An overview of OpenMP (talk) 10:35   Coffee 10:50    OpenMP programming and execution model (talk+practical) 12:30   Lunch 13:30    OpenMP worksharing directives (talk+practical)  15:00   Coffee 15:15    More OpenMP (talk+practical) 16:00    Summary (talk) + Q & A 16:30   End of first day 2nd day 08:45  Join online 09:00    How to write correct OpenMP programs (talk) 09:30    Intel Inspector XE (demo+practical) 10:30   Coffee 10:45    How to write efficient OpenMP programs (talk) 11:00    Heat equation (talk+practical) 12:30   Lunch 13:30    Heat equation (practical — continued) 14:00    Vectorization with OpenMP (talk) 14:25    Thread affinity in OpenMP (talk) 15:00   Coffee 15:15     The taskloop directive (talk+practical) 16:30   End of second day (course) https://events.prace-ri.eu/event/1250/ 2021-10-28 06:45:00 UTC 2021-10-29 14:30:00 UTC [] [] [] workshops_and_courses [] []
  • [ONLINE] CUDA 4 Dummies @ VSC Vienna

    9 - 10 December 2021

    [ONLINE] CUDA 4 Dummies @ VSC Vienna https://tess.elixir-europe.org/events/online-cuda-4-dummies-vsc-vienna-b6d97f62-473e-44ea-9a0c-100c00d4a5b9 This course will be delivered as an ONLINE COURSE (using Zoom) for remote participation. REGISTRATION is strictly NECESSARY since the details to access the online course will be provided to the registered and accepted attendees only. Registration follows the "first come – first served" principle. For the online course we can take a maximum of 40 persons. Additional registrations will be added to a waiting list. Furthermore, accepted participants will be contacted a few days before the course and asked to do a short PRE-ASSIGNMENT that has to be completed before the course starts. The HANDS-ON LABS will be done on VSC-3, all participants will get a training user account at the VSC for the duration of the course. Overview CUDA is the standard API for code development targeting the GPU and a number of impressive examples have already been given in diverse areas from particle physics to computational biology. CUDA allows simple extensions of standard C/C++/Fortran code with GPU-specific functions. In this way thousands of cores available on the GPU can be leveraged to work in parallel and thus carry out significant fractions of the computational workload on the device rather than the CPU. There is also a vast set of auxiliary tools available to the developer including libraries, code templates, building blocks, analysis tools, developmental frameworks and in general a vivid community making up the CUDA Developer Zone. It is often for this multifaceted support environment that the interested beginner is feeling overwhelmed and unsettled about which particular first steps should best be taken to gain a straightforward introduction into the subject. For exactly this reason the present course is offering a systematic step-by-step introduction into GPU computing from the perspective of the newcomer. Basic design principles will be established, central programming techniques acquired and a number of state-of-the-art workflows examined that efficiently employ the GPU and are frequently used in scientific computing. This course provides lectures, demos, and hands-on labs. This course is a PRACE training event. It is organized by the VSC Research Center, TU Wien. Agenda & Content (tentative) 1st day – 9 December 2021 Basic design principles, getting started with simple CUDA routines, mastering the CUDA environment; Memory Hierarchy with a focus on CUDA managed unified memory; Practical exercise: all the pitfalls one can run into when porting a dummy code to CUDA. 08:45 - 09:00    join online 09:00 - 13:00    lectures and hands-on labs (with breaks: 10:45-11:00) 13:00 - 14:00    lunch break 14:00 - 17:00    lectures and hands-on labs (with breaks: 15:20-15:35) 2nd day – 10 December 2021 Selected examples to demonstrate CUDA capabilities and GPU operation, e.g., bandwidthTest, simpleAssert, simpleStreams, etc.; CUDA libraries, CUBLAS, CUSolver, considering numerical precision, gaining insight into the runtime; Practical exercise: CUDA in everyday life (of a scientist). 08:45 - 09:00    join online 09:00 - 13:00    lectures and hands-on labs (with breaks: 10:45-11:00) 13:00 - 14:00    lunch break 14:00 - 17:00    lectures and hands-on labs (with breaks: 15:20-15:35) https://events.prace-ri.eu/event/1251/ 2021-12-09 07:05:00 UTC 2021-12-10 16:00:00 UTC [] [] [] workshops_and_courses [] []