BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//The NCC Netherlands - ECPv6.15.18//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://eurocc-netherlands.nl
X-WR-CALDESC:Events for The NCC Netherlands
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
TZOFFSETFROM:+0000
TZOFFSETTO:+0000
TZNAME:UTC
DTSTART:20220101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=UTC:20230502T080000
DTEND;TZID=UTC:20230505T170000
DTSTAMP:20260404T071417
CREATED:20230411T135540Z
LAST-MODIFIED:20230411T135540Z
UID:385-1683014400-1683306000@eurocc-netherlands.nl
SUMMARY:EuroCC-AMD Workshop: Developing HPC Applications with AMD GPUs
DESCRIPTION:AMD\, in collaboration with CASTIEL-EuroCC\, will be giving a workshop on developing HPC applications for AMD GPUs. The workshop will begin with a developer’s view of the AMD system GPUS and associated hardware. Then it will cover a few different programming options including HIP and OpenMP. There will be some presentations on obtaining the best performance for your application. The workshop will then introduce some of the available AMD profiling and debugging tools. \nThe workshop is open for CASTIEL and EuroCC due to the limited amount of places and it targets application developers with an understanding of recent HPC computing hardware and computing environments. \nNo prior GPU programming experience is necessary. \nThe content will be about 30%/40%/30% Beginner/Intermediate/Advanced. \n  \n  \n2 May 2023Time                     Topic13:00 – 13:15     CASTIEL Intro13:15 – 13:25     AMD Presentation Roadmap13:25 – 13:45     Introduction to the System for Exercises13:45 – 14:00     Introduction to AMD Architecture14:00 – 14:15     Break14:15 – 15:15     Introduction to HIP15:15 – 15:45     HIP Exercises15:45 – 16:00     HIPFort16:00 – 16:15     Break16:15 – 16:40     Porting applications to HIP16:40 – 16:55     Porting exercises16:55 – 17:00     Wrap-up \n3 May 2023Time                    Topic13:00 – 13:50     Getting Started with OpenMP® Offload Applications on AMD Accelerators13:50 – 14:30     OpenMP® exercises14:30 – 14:45     Break14:45 – 15:00     OpenACC15:00 – 16:00     Advanced OpenMP® offloading16:00 – 16:10     Break16:10 – 16:40     Advanced OpenMP® Exercises16:40 – 16:50     Mixing HIP and OpenMP®16:50 – 16:55     Infinity Hub/HPC Community16:55 – 17:00     Wrap-up \n4 May 2023Time                     Topic13:00 – 13:40     Kokkos13:40 – 13:55     Kokkos Exercises13:55 – 14:05     Break14:05 – 14:30     AMD Communication Fabrics and GPU-Aware MPI14:30 – 14:45     GPU Aware Exercises14:45 – 14:55     Break14:55 – 15:35     AMD Node Memory Model15:35 – 15:50     Memory Model Exercises15:50 – 16:00     Break16:00 – 16:40     Affinity — Placement\, Ordering and Binding16:40 – 16:55     Affinity exercises16:55 – 17:00     Wrap-up \n5 May 2023Time                     Topic13:00 – 13:20     Introduction to Rocprof13:20 – 13:35     Rocprof Exercises13:35 – 14:00     Introduction to Omnitrace14:00 – 14:15     Omnitrace Exercises14:15 – 14:25     Break14:25 – 15:25     Roofline15:25 – 15:55     Introduction to Omniperf15:55 – 16:10     Exercises – Omniperf16:10 – 16:20     Break16:20 – 16:55     Debuggers – rocgdb16:55 – 17:00     Wrap-up \nYou need to register in advance for this workshop.Before the event\, you will receive a link from Webex to join the meeting. \n  \nIf you would like to participate in the workshop\, please register in advance by Monday 24 April\, EoD: https://www.eurocc-access.eu/event/eurocc-amd-workshop-developing-hpc-applications-with-amd-gpus/
URL:https://eurocc-netherlands.nl/calendar/eurocc-amd-workshop-developing-hpc-applications-with-amd-gpus/
CATEGORIES:Training
ATTACH;FMTTYPE=image/png:https://eurocc-netherlands.nl/wp-content/uploads/2023/04/AMD-workshop-eurocc.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20230516T093000
DTEND;TZID=UTC:20230516T170000
DTSTAMP:20260404T071417
CREATED:20230425T132003Z
LAST-MODIFIED:20230425T132540Z
UID:393-1684229400-1684256400@eurocc-netherlands.nl
SUMMARY:Machine Learning on HPC systems (1 day)
DESCRIPTION:You want to train a neural network for your research project\, and have just gotten access to a high performance cluster with a lot of powerful hardware. Great! But\, how can you make sure that you’ll use these (expensive!) resources effectively? In this course\, you will learn how to get the most results out of the computational budget you were granted. \n  \n\n\n\nIn this course you will learn:\n\nHow to set up your software environment\, and why the preinstalled software modules are useful;\nHow the file I/O might limit your training speed\, and how to overcome that;\nAbout the technical capabilities of modern day CPUs and GPUs (reduced precision datatypes\, vector/matrix instructions);\nHow to find bottlenecks in your code through creating a (PyTorch) profile;\nHow to use multiple CPUs or GPUs in a single training (parallel computing for deep learning).\n\nWho?\nMachine Learning researchers whose requirements for training their neural networks have outgrown their local computer\, and are using or planning to use a high performance computing cluster (such as Snellius) to train their models. \nPrerequisites\n\nBasic knowledge in PyTorch\, TensorFlow or a similar framework;\nBasic knowledge on Python programming. Some experience in using Jupyter notebooks is desireable\, but not essential;\nBasic knowledge in using a high performance computing cluster (see our course ‘Introduction to cluster and supercomputing);\nSpecifically: know how to submit a job\, and how to interact with the module environment.\n\n\n\n\nSign up: https://events.surf.nl/kort4/open/8e6e2802-7165-479f-ae4e-be242555961e  *Please note\, registration closes May 15! 
URL:https://eurocc-netherlands.nl/calendar/machine-learning-on-hpc-systems-1-day/
LOCATION:SURF Amsterdam\, Science Park 140\, Amsterdam\, 1098 XG\, Netherlands
CATEGORIES:Training
ATTACH;FMTTYPE=image/jpeg:https://eurocc-netherlands.nl/wp-content/uploads/2023/04/Machine-Learning-Training-16-may.jpeg
END:VEVENT
END:VCALENDAR