BEGIN:VCALENDAR VERSION:2.0 PRODID:-//CERN//INDICO//EN BEGIN:VEVENT SUMMARY:GPU Programming Workshop DTSTART:20260608T070000Z DTEND:20260611T140000Z DTSTAMP:20260528T171500Z UID:indico-event-364@events.hpc-portal.eu DESCRIPTION:Overview\nIn this 4-day online workshop you will learn how to accelerate your applications with OpenACC\, OpenMP\, Kokkos\, CUDA C++ and CUDA Python on NVIDIA GPUs.\nThe lectures are interleaved with many hands -on sessions using\n\nThe JupyterHub at NHR@FAU to access the Alex cluster (Day 1 and Day 2)\,\nDLI Jupyter Notebooks to access a fully configured G PU-accelerated workstation in the cloud (Day 3 and Day 4).\n\nThe workshop is co-organised by Leibniz Supercomputing Centre (LRZ)\, Erlangen Nati onal High Performance Computing Center (NHR@FAU) and NVIDIA Deep Learnin g Institute (DLI). NVIDIA DLI offers hands-on training for developers\, d ata scientists\, and researchers looking to solve challenging problems wit h deep learning.\nDay 1 (OpenACC) and Day 2 (OpenMP and Kokkos) are offere d outside of the NVIDIA DLI programme.\nAgenda\n1st day: Fundamentals of A ccelerated Computing with OpenACC\nLearn the basics of OpenACC\, a high-le vel programming language for programming on GPUs. This lecture is for anyo ne with some C/C++ of Fortran experience who is interested in accelerating the performance of their applications beyond the limits of CPU-only progr amming. In this lecture\, you’ll learn:\n\nHow to profile and optimise y our CPU-only applications to identify hot spots for acceleration\nHow to u se OpenACC directives to GPU accelerate your codebase\nHow to optimise dat a movement between the CPU and GPU accelerator\n\nUpon completion\, you'll be ready to use OpenACC to GPU accelerate CPU-only applications.\n \n2nd day: Fundamentals of Accelerated Computing with OpenMP and Kokkos\nMorni ng session: Introduction to GPU programming using OpenMP directives. The l ecture covers the OpenMP offloading model\, including parallelism\, data m apping\, and memory management.\nAfternoon session: GPU programming with K okkos\, a modern C++ library for performance-portable parallel computing. Participants will learn to implement algorithms efficiently on GPUs\, leve rage parallelism\, and optimise memory access to maximise performance.\n  \n3rd day: Fundamentals of Accelerated Computing with Modern CUDA C++\nThi s lecture provides a comprehensive introduction to general-purpose GPU pro gramming with CUDA. You'll learn how to write\, compile\, and run GPU-acce lerated code\, leverage CUDA core libraries to harness the power of massiv e parallelism provided by modern GPU accelerators\, optimize memory migrat ion between CPU and GPU\, and implement your own algorithms. At the end of the lecture\, you'll have access to additional resources to create your o wn GPU-accelerated applications.\n \n4th day: Fundamentals of Accelerate d Computing with CUDA Python\nThis lecture explores how to use Numba — t he just-in-time\, type-specialising Python function compiler — to accele rate Python programs to run on massively parallel NVIDIA GPUs. You’ll le arn how to:\n\nUse Numba to compile CUDA kernels from NumPy universal func tions (ufuncs)\nUse Numba to create and launch custom CUDA kernels\nApply key GPU memory management techniques\n\nUpon completion\, you’ll be able to use Numba to compile and launch CUDA kernels to accelerate your Python applications on NVIDIA GPUs.\nImportant information\nAfter you are accept ed\, please create an account under https://learn.nvidia.com/join\nEnsure your laptop / PC will run smoothly by going to http://websocketstest.com /Make sure that WebSockets work for you by seeing under Environment\, WebS ockets is supported and Data Receive\, Send and Echo Test all check Yes un der WebSockets (Port 80).If there are issues with WebSockets\, try updatin g your browser.\nNVIDIA Deep Learning Institute\nThe NVIDIA Deep Learning Institute delivers hands-on training for developers\, data scientists\, and engineers. The program is designed to help you get started with traini ng\, optimising\, and deploying neural networks to solve real-world proble ms across diverse industries such as self-driving cars\, healthcare\, onli ne services\, and robotics.\nPrerequisites\nDay 1:\n\nBasic C/C++ or Fortr an competency\, including familiarity with variable types\, loops\, condit ional statements\, functions\, and array manipulations.\nNo previous knowl edge of OpenACC programming is assumed.\n\nDay 2:\n\nBasic C++ competency\ , including familiarity with variable types\, loops\, conditional statemen ts\, functions\, and array manipulations.\nNo previous knowledge of OpenMP and Kokkos programming is assumed.\n\nDay 3: \n\nBasic C++ competency\, including familiarity with lambda expressions\, loops\, conditional statem ents\, functions\, standard algorithms and containers.\nNo previous knowle dge of CUDA programming is assumed.\n\nDay 4:\n\nBasic Python competency\, including familiarity with variable types\, loops\, conditional statement s\, functions\, and array manipulations.\nNumPy competency\, including the use of ndarrays and ufuncs.\nNo previous knowledge of CUDA programming is required.\n\nHands-On\nThe lectures are interleaved with many hands-on se ssions using\n\nThe JupyterHub at NHR@FAU to access the Alex cluster (Day 1 and Day 2)\,\nDLI Jupyter Notebooks to access a fully configured GPU-acc elerated workstation in the cloud (Day 3 and Day 4).\n\nLanguage\nEnglish\ nLecturers\nDr. Momme Allalen\, Dr. Sajjad Azizi\, Dr. Volker Weinberg ( all LRZ)\, Dr. Sebastian Kuckuk (NHR@FAU)\nPrices and Eligibility\nThe course is open and free of charge for academic participants from the Membe r States of the European Union (EU) and Associated Countries to the Horiz on 2020 programme.\nRegistration\nPlease register with your official e-mai l address to prove your affiliation.\nWithdrawal Policy\nSee Withdrawal\n Legal Notices\nFor registration for LRZ courses and workshops we use the s ervice edoobox from Etzensperger Informatik AG (www.edoobox.com). Etzenspe rger Informatik AG acts as processor and we have concluded a Data Processi ng Agreement with them.\n\nhttps://events.hpc-portal.eu/event/364/ LOCATION:Online URL:https://events.hpc-portal.eu/event/364/ END:VEVENT END:VCALENDAR