BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:Europe/Stockholm X-LIC-LOCATION:Europe/Stockholm BEGIN:DAYLIGHT TZOFFSETFROM:+0100 TZOFFSETTO:+0200 TZNAME:CEST DTSTART:19700308T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:+0200 TZOFFSETTO:+0100 TZNAME:CET DTSTART:19701101T020000 RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20230831T095754Z LOCATION:Flüela DTSTART;TZID=Europe/Stockholm:20230626T163000 DTEND;TZID=Europe/Stockholm:20230626T183000 UID:submissions.pasc-conference.org_PASC23_sess142@linklings.com SUMMARY:MS2B - Performance Portability Solutions beyond C++ to Support Fut ure Workflows (Part 2/2) DESCRIPTION:Minisymposium\n\nComputing at large scales has become extremel y challenging due to increasing heterogeneity in both hardware and softwar e. A positive feedback loop exists where more scientific insight leads to more complex solvers which in turn need more computational resource. More and more scientific workflows need to tackle a range of scales and use mac hine learning (ML) and artificial intelligence (AI) intertwined with more traditional numerical modeling methods, placing more demands on computatio nal platforms. These constraints indicate a need to fundamentally rethink the way computational science is done and the tools that are needed to ena ble these complex workflows. It is not obvious that current set of C++ bas ed solutions will suffice, or that relying exclusively upon C++ is the bes t option, especially because several newer languages and boutique solution s offer more robust design features to tackle the challenges of heterogene ity. This two part minisymposium will include presentations about language s and heterogeneity solutions that are not tied to C++, and offer options beyond template metaprogramming and parallel-for for performance and porta bility. One slot will be reserved for open discussion and exchange of idea s.\n\nPerformance Portability Using a Tool-Chain to Exploit Separation of Concerns\n\nMulti-scale, multi-physics scientific and engineering simulati on codes take years to develop and optimize. At the same time effective ut ilization of high performance computing (HPC) resources has always been a balancing act between portability and performance. Increasing heterogeneit y and ongoing de...\n\n\nAnshu Dubey (Argonne National Laboratory, Univers ity of Chicago); Jared O'Neal (Argonne National Laboratory); Johann Rudi ( Virginia Tech, Argonne National Laboratory); Mohamed Wahib (RIKEN); Tom Kl osterman (Argonne National Laboratory); and Klaus Weide (University of Chi cago, Argonne National Laboratory)\n---------------------\nData-Centric Py thon: Bridging Productivity and Performance via Data Movement Minimization \n\nComputational scientists are migrating towards high-productivity langu ages for rapid prototyping and reproducible experiment sharing. Specifical ly, Python is becoming the language of choice for several fields, partly d riven by the attention from the Machine Learning community. However, produ ctivity ...\n\n\nTal Ben-Nun (Lawrence Livermore National Laboratory)\n--- ------------------\nGenerating Optimal HPC Code with Machine Learning\n\nF ollowing the deep learning revolution started approximately a decade ago, experts in both fields of compilers and machine learning have approached t he problem of generating code using ML methods. Results include, e.g., ML models generating (often suboptimal and sometimes incorrect) code or non-n egl...\n\n\nEmil Vatai (RIKEN)\n---------------------\nGeneral Discussion\ n\nThis presentation slot will be used to discuss the merits of performanc e portability solutions presented in this mini-symposium and how they stac k up against more prevalent C++ solutions.\n\n\nAnshu Dubey (Argonne Natio nal Laboratory, University of Chicago)\n\nDomain: Computer Science, Machin e Learning, and Applied Mathematics 
\n\nSession Chair: Anshu Dubey (Argon ne National Laboratory, University of Chicago) END:VEVENT END:VCALENDAR