BEGIN:VCALENDAR VERSION:2.0 PRODID:icalendar-ruby CALSCALE:GREGORIAN BEGIN:VEVENT DTSTAMP:20260703T190421Z UID:103e69d8-fc62-43d1-83c5-d30a677dce58 DTSTART:20220208T090000Z DTEND:20220210T150000Z DESCRIPTION:With the increasing prevalence of multicore processors\, shared -memory programming models are essential. OpenMP is a popular\, portable\, widely supported\, and easy-to-use shared-memory model. \n\nSince its adv ent in 1997\, the OpenMP programming model has proved to be a key driver b ehind parallel programming for shared-memory architectures.  Its powerful and flexible programming model has allowed researchers from various domai ns to enable parallelism in their applications.  Over the more than two d ecades of its existence\, OpenMP has tracked the evolution of hardware and the complexities of software to ensure that it stays as relevant to today ’s high performance computing community as it was in 1997.\n\nThis works hop will cover a wide range of  topics\, reaching from the basics of Open MP programming using the "OpenMP Common Core" to really advanced topics. D uring each day lectures will be mixed with hands-on sessions.\n\nTentative Agenda\n\n\n \n \n \n  \n \n \n Day 1\n \n \n Day 2 \n \n \n Day 3\n \n \n \n \n 09:00-10:30 CET\n \n \n Intr oduction to OpenMP 1\n \n \n Tasking 1\n\n \n Tasking Intro\n Lab 1\n \n \n \n GPUs\n\n \n OpenMP for Compute Accelerator s\n \n \n \n \n \n 10:45-12:15 CET\n \n \n Hands-on: Intro duction to OpenMP\n \n \n Tasking 2\n\n \n Taskloop\n Depend encies\n Cancellation\n Lab 2\n \n \n \n Tools for Perf. and Correctness\n\n \n VI-HPS Tools for Performance\n VI-HPS Tools fo r Correctness\n \n \n \n \n \n 13:00-14:45 CET\n \n \n Int roduction to OpenMP 2\n \n  \n \n \n Host Perf.: SIMD\n\n \n Vectorisation\n Lab 3\n \n \n \n Misc. OpenMP 5.1 Features\n \n \n DOACROSS Loops\n \n \n \n \n \n 15:00-16:00 CET\n \ n \n Hands-on: Introduction to OpenMP\n \n \n Host Perf.: NUMA\n \n \n Memory Access\n Task Affinity\n Memory Management\n La b 4\n \n \n \n Roadmap / Outlook\n\n \n Open Discussion\n OpenMP 5.1 and beyond\n \n \n \n \n\n\nDay 1\n\nThe first day will co ver basic parallel programming with OpenMP. \n\nMost OpenMP programmers us e only around 21 items from the specification. We call these the “OpenMP Common Core”. By focusing on the common core on the first day\, we make OpenMP what it was always meant to be: a simple API for parallel applicat ion programmers.\n\nIn this hands-on tutorial\, students use active learni ng through a carefully selected set of exercises\, to master the Common Co re and learn to apply it to their own problems. \n\nDay 2 and 3\n\nDay 2 a nd 3 will cover advanced topics like:\n\n\n Mastering Tasking with OpenMP\ , Taskloops\, Dependencies and Cancellation\n Host Performance: SIMD / Vec torization\n Host Performance: NUMA Aware Programming\, Memory Access\, Ta sk Affinity\, Memory Management\n Tool Support for Performance and Correct ness\, VI-HPS Tools \n Offloading to Accelerators\n Other Advanced Feature s of OpenMP 5.1\n Future Roadmap of OpenMP\n\n\nDevelopers usually find Op enMP easy to learn. However\, they are often disappointed with the perform ance and scalability of the resulting code. This disappointment stems not from shortcomings of OpenMP but rather with the lack of depth with which i t is employed. The lectures on Day 2 and Day 3 will address this critical need by exploring the implications of possible OpenMP parallelization stra tegies\, both in terms of correctness and performance. \n\nWe cover taskin g with OpenMP and host performance\, putting a focus on performance aspect s\, such as data and thread locality on NUMA architectures\, false sharing \, and exploitation of vector units. Also tools for performance and correc tness will be presented.\n\nCurrent trends in hardware bring co-processors such as GPUs into the fold. A modern platform is often a heterogeneous sy stem with CPU cores\, GPU cores\, and other specialized accelerators. Open MP has responded by adding directives that map code and data onto a device \, the target directives. We will also explore these directives as they a pply to programming GPUs.\n\nOpenMP 5.1 features will be highlighted and t he future roadmap of OpenMP will be presented.\n\nAll topics are accompani ed with extensive case studies and we discuss the corresponding language f eatures in-depth.\n\nTopics might be still subject to change.\n\nFor the h ands-on sessions participants need to use their own laptops or systems wit h a C/C++ or Fortran compiler supporting at least OpenMP 4.5 installed (se e https://www.openmp.org/resources/openmp-compilers-tools/).\n\nThe course is organized as a PRACE training event by LRZ in collaboration with the O penMP ARB and RWTH Aachen.\n\nPrerequisites\n\nBasic C/C++ or Fortran know ledge. Basic OpenMP knowledge is useful for Day 2 and Day 3 of the Worksho p\, but will also be provided on the first day.\n\nHands-On\n\nFor the han ds-on sessions participants need to use their own laptops or systems with a C/C++ or Fortran compiler supporting at least OpenMP 4.5 installed (see https://www.openmp.org/resources/openmp-compilers-tools/).\n\nContent Leve l\n\nThe content level of the course is broken down as:\n\n\n \n \n \n Beginner's content:\n \n \n 33.3%\n \n \n \n \n Intermedia te content:\n \n \n 33.3%\n \n \n \n \n Advanced content:\n \n \n 33.3%\n \n \n \n \n Community-targeted content:\n \n \n 0%\n \n \n \n\n\nLanguage\n\nEnglish\n\nLecturers\n\nDr.-Ing. M ichael Klemm (OpenMP ARB\, AMD)\, Dr. Christian Terboven (RWTH Aachen Univ ersity)\n\nDr.-Ing. Michael Klemm is part of the HPC Center of Excellence at AMD. His focus is on High Performance &\; Throughput Computing Enabl ing. He obtained an M.Sc. in Computer Science in 2003. He received a Docto r of Engineering degree (Dr.-Ing.) in Computer Science from the Friedrich- Alexander-University Erlangen-Nuremberg\, Germany\, in 2008. His research focus was on compilers and runtime optimizations for distributed systems. His areas of interest include compiler construction\, design of programmin g languages\, parallel programming\, and performance analysis and tuning. He is CEO of the OpenMP Language Committee\, where he is also leading the group on the OpenMP Error Model efforts.\n\nDr. Christian Terboven is a se nior scientist and leads the HPC group at RWTH Aachen University. His rese arch interests center around Parallel Programming and related Software Eng ineering aspects. Dr. Terboven has been involved in the Analysis\, Tuning and Parallelization of several large-scale simulation codes for various ar chitectures. He is responsible for several research projects in the area o f programming models and approaches to improve the productivity and effici ency of modern HPC systems. He is further co-author of the new book “Usi ng OpenMP – The Next Step“\, https://www.openmp.org/tech/using-openmp- next-step/.\n\nPrices and Eligibility\n\nThe course is open and free of ch arge for people from academia and industry from the Member States (MS) of the European Union (EU) and Associated/Other Countries to the Horizon 2020 programme.\nhttps://events.prace-ri.eu/event/1279/ SUMMARY:[ONLINE] OpenMP Programming Workshop @ LRZ URL;VALUE=URI:https://events.prace-ri.eu/event/1279/ END:VEVENT END:VCALENDAR