BEGIN:VCALENDAR VERSION:2.0 PRODID:Linklings LLC BEGIN:VTIMEZONE TZID:America/Chicago X-LIC-LOCATION:America/Chicago BEGIN:DAYLIGHT TZOFFSETFROM:-0600 TZOFFSETTO:-0500 TZNAME:CDT DTSTART:19700308T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU END:DAYLIGHT BEGIN:STANDARD TZOFFSETFROM:-0500 TZOFFSETTO:-0600 TZNAME:CST DTSTART:19701101T020000 RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20260522T150118Z LOCATION:D220 DTSTART;TZID=America/Chicago:20181111T090000 DTEND;TZID=America/Chicago:20181111T173000 UID:submissions.supercomputing.org_SC18_sess160@linklings.com SUMMARY:Women in HPC: Diversifying the HPC Community DESCRIPTION:From Message Passing to PGAS\n\nThe continuous increase in the number of nodes across the whole system in supercomputers makes Remote-me mory-access (RMA) a good alternative for improving applications' performan ce on these machines. However, in order for existing applications to benef it from it, they have to change their code to in...\n\n\nHadia Ahmed and S cott B. Baden (Lawrence Berkeley National Laboratory)\n------------------- --\nIn Situ Performance Analysis of Event-driven Simulations to Support th e Codesign of Extreme-Scale Systems\n\nParallel discrete event simulation (PDES) is a productive and cost-effective tool in exploring the design spa ce of high performance computing (HPC) system architectures. In discrete event simulation, the modeled entities (e.g., network router) interact thr ough the exchange of timestamped messages. ...\n\n\nCaitlin J. Ross (Renss elaer Polytechnic Institute)\n---------------------\nTitle: Distributed Me mory Fast Fourier Transforms in the Exascale Era\n\nIn this project, we pl an to improve the implementation of the Fast Fourier Transform (FFT) on em erging high-performance computing architectures and galvanize the internat ional community of experts to benchmark our and their efforts, leading the faster and more efficient adoption of the best FFT soft...\n\n\nSamar Ase eri (King Abdullah University of Science and Technology)\n---------------- -----\nFLAME GPU: Complex System Simulation Framework\n\nAgent based model ing can be used to represent complex systems at differing biological scale s\n\nFLAME GPU (Flexible Large Scale Agent Based Modelling Environment) :\ n• Aim: High performance agent based simulation without writing lots of GP U code\n• Technique: Agents are state machines with memory (commu...\n\n\n Mozhgan Kabiri Chimeh (University of Sheffield)\n---------------------\nWo rkshop Lunch (on your own)\n---------------------\nA Black Woman’s Sojourn in High Performance Computing: Recovering Lost History\n\nIn 1797, Isabel la Bomfree was born a slave. She escaped to freedom in 1827 and changed he r name to Sojourner Truth in 1843 to indicate that she would travel far an d wide to tell people what was right. She gave her famous “Ain’t I a Woman ?” speech in 1851 to highlight the strengths and vulnerabilitie...\n\n\nRu by Mendenhall (University of Illinois)\n---------------------\nOptimizatio n of a Lattice Boltzmann Program\n\nIn this project, I optimized a Lattice Boltzmann (LB) program developed internally by the digital rock team at B P to simulate single-phase flow and predict permeability of porous rocks. LB algorithm is simple to formulate and implement, and it is easily parall elizable on computer clusters. My work i...\n\n\nMuhong Zhou (Numerical Al gorithms Group, BP Center For High Performance Computing)\n--------------- ------\nWorkshop Afternoon Break\n---------------------\nPanel Discussion – Best Practices from Organizations on Improving Workplace Diversity.\n\nT he panel discussion will address strategies to improve diversity and inclu sion at workplaces.\n\n\nChristine Cuicchi (US Department of Defense)\n--- ------------------\nBest Practices from Organizations on Improving Workpla ce Diversity\n\nPurdue Research Computing has been a proven leader in cult ivating future STEM professionals for over a decade and creating a diverse workforce. We have built a pro-diversity and inclusive workplace culture that embraces a variety of academic degrees, generations, demographics, ex pertise, and experie...\n\n\nElizabett Hillery, Claire Stirm, Carolyn Elli s, and Laura Theademan (Purdue University)\n---------------------\nChallen ges of Performance Portability for Fortran Unstructured Mesh Codes\n\nWhat pathways exist for Fortran performance portability to exascale? Fortran-b ased codes present different challenges for performance portability and pr oductivity. Ideally, we want to develop one codebase that can run on many different HPC architectures. In reality, each architecture has its own idi ...\n\n\nAbigail Hsu (Stony Brook University, Los Alamos National Laborato ry) and David H. Neill Asanza (Grinnell College, Los Alamos National Labor atory)\n---------------------\nA Deferred Correction Coupling Strategy fo r Cosmological Simulations\n\nWe present a strategy for coupling the equat ions of baryonic gas dynamics to those for radiative heating and cooling i n the context of Nyx, an N-body + gas dynamics code for large-scale cosmol ogical simulations. The radiative and advective processes have differing t ime scales. To avoid Strang splitt...\n\n\nJean M. Sexton (Lawrence Berkel ey National Laboratory)\n---------------------\nWorkshop Morning Break\n-- -------------------\nError Analysis in HPC Applications Using Algorithmic Differentiation\n\nComputer applications running on supercomputers are use d to solve critical problems. These systems are expected to perform tasks not just quickly, but also correctly. Various factors that can affect cor rectness of programs include faults, reduced precision, lossy data reducti on, iteration and trunc...\n\n\nHarshitha Menon (Lawrence Livermore Nation al Laboratory)\n---------------------\nDeep Learning: Extrapolation Tool for Computational Nuclear Physics\n\nThe goal of nuclear theory is to unde rstand how nuclei arise from interacting nucleons based on the underlying theory of the strong interactions, quantum chromodynamics (QCD). The inter actions among the nucleons inside a nucleus are dominated by the strong in teraction, which is non-perturbative in t...\n\n\nGianina Alina Negoita (I owa State University)\n---------------------\nEfficient Application of Low Mach Number Hydrodynamics Code to Stellar Flows\n\nAstrophysical phenomen a that occur in the low Mach number regime are often computationally expen sive because the time step is constrained by both the characteristic fluid velocity and, by comparison, the much larger speed of sound. In addition, astrophysical flows are generally highly turbulent, and...\n\n\nDuoming F an (Lawrence Berkeley National Laboratory)\n---------------------\nThe Mov ement toward HPC Inclusivity: Achieving On-Demand Accessibility of High Pe rformance Computing (HPC) through Ephemeral Projects Utilizing the Alces G ridware Project\n\nThis presentation covers off the adoption of public clo ud over the past two years. Using five public case studies, we will discu ss how moving into cloud is not an 'all or nothing' option and demonstrate how adding cloud to an HPC solution invites in more opportunities for HPC to grow. But jumping ...\n\n\nCristin Merritt and Wil Mayers (Alces Flig ht Limited)\n---------------------\nHot Topics Discussion II: Thriving at Work\n\nThe session will resume hot topic presentations on the following t opics:\n- Effectively responding to discrimination -- Lorna Rivera, Georg ia Tech \n- Effective workplace communication -- Lucy Nowell, DoE Office of Science\n- Coming out as a transgender woman at a major HPC center -- Carissa Holoha...\n\n\nLorna Rivera (Georgia Institute of Technology), Lu cy Nowell (US Department of Energy Office of Advanced Scientific Computing Research), and Carissa Holohan (Argonne National Laboratory)\n----------- ----------\nCharacterization of the Impact of Soft Errors on Iterative Met hods\n\nSoft errors caused by transient bit flips can lead to silent data corruption which can significantly impact an application’s behavior. When a transient bit flip affects a hardware component, the application is said to be impacted by a soft error. When a soft error escapes hardware detect ion and impa...\n\n\nBurcu Mutlu (Pacific Northwest National Laboratory, P olytechnic University of Catalonia); Gokcen Kestor (Oak Ridge National Lab oratory); Osman Unsal (Barcelona Supercomputing Center); and Sriram Krishn amoorthy (Pacific Northwest National Laboratory)\n---------------------\nP lasma Meets Portability: A Journey to Performance Portability in a Particl e-in-Cell Code\n\nIn our journey to exascale, it is imperative that applic ations not only run efficiently, but do so on a wide variety of platforms at a reasonable cost. Codes must thus deal with a explosion of diversity a mong emerging architectures while avoiding a corresponding explosion in co de size. This requires...\n\n\nJoy Kitson (University of Delaware, Los Ala mos National Laboratory) and Stephen Lien Harrell (Purdue University)\n--- ------------------\nUsing a Robust Metadata Management System to Accelerat e Scientific Discovery at Extreme Scales\n\nLarge-scale scientific simulat ions are an important tool for scientific discovery. In recent years, ther e has been a rapid growth in the amount of data output by these simulation s. Extended runs of simulations such as XGC edge plasma fusion can easily generate datasets in the terabyte to petabyte ra...\n\n\nMargaret Lawson ( University of Illinois, Sandia National Laboratories)\n------------------- --\nLarge-Scale PDE-Constrained Optimization\n\nOptimization of time-depen dent PDE-constrained optimization problems is extremely challenging from a computational perspective. Presume one forward simulation of a differenti al equation with N degrees of freedom, advancing in time for M timesteps r equires a time to solution T. In this case optimizin...\n\n\nOana Marin (A rgonne National Laboratory)\n---------------------\nStudy of Performance V ariability on Dragonfly Systems\n\nDragonfly networks are being widely ado pted in high-performance computing systems. On these networks, however, in terference caused by resource sharing can lead to significant network cong estion and performance variability. On a shared network, different job pla cement policies lead to different traff...\n\n\nXin Wang (Illinois Institu te of Technology)\n---------------------\nKinetic Simulations of Plasma Tu rbulence Using the Discontinuous Galerkin Finite Element Method\n\nWe use an advanced computational framework called Gkeyll to simulate plasma turbu lence in the edge of magnetic confinement fusion experiments. In these exp eriments, charged particles gyrate quickly around magnetic field lines. Wh en time scales of interest are much greater than the gyration period an... \n\n\nTess Bernard (University of Texas), Ammar Hakim and Gregory Hammett (Princeton Plasma Physics Laboratory), Eric Shi (Lawrence Livermore Nation al Laboratory), Manaurae Francisquez (Princeton Plasma Physics Laboratory) , and James Juno (University of Maryland)\n---------------------\nHot Topi cs Discussion I: Thriving at Work\n\nThe session will be focused on hot to pic presentations including the following topics:\n- Being a diversity al ly—what it means and how you can help by Linda Akli, Southeastern Universi ties Research Association\n- Telecommuting boss - being a successful remo te employee by Maytal Dahan, TACC\n- Leader...\n\n\nLinda Akli (Southeast ern Universities Research Association (SURA)), Maytal Dahan (University of Texas), and Laura Schulz (Leibniz Supercomputing Centre)\n--------------- ------\nOptimizing Python Data Processing for the DESI Experiment on the N ERSC Cori Supercomputer\n\nThe goal of the Dark Energy Spectroscopic Instr ument (DESI) experiment is to better understand dark energy by making the most detailed 3D map of the universe to date. The images obtained each nig ht over a period of 5 years starting in 2019 will be sent to the NERSC Cor i supercomputer for processing ...\n\n\nLaurie A. Stephey and Rollin C. Th omas (Lawrence Berkeley National Laboratory, National Energy Research Scie ntific Computing Center (NERSC)) and Stephen J. Bailey (Lawrence Berkeley National Laboratory)\n---------------------\nIntroduction - Women in HPC: Diversifying the HPC Community\n\nThe ninth international Women in HPC wor kshop will be held at SC18, Dallas, USA. Following the overwhelming succes s of the WHPC workshop at SC17, we will focus on the following topics this year:\n- Building workplace resilience and maintaining well-being, while managing work stress.\n- Being part of t...\n\n\nMisbah Mubarak (Argonne N ational Laboratory); Elsa Gonsiorowski (Lawrence Livermore National Labora tory); and Toni Collis (Appentra Solutions, Women in High Performance Comp uting)\n---------------------\nDeveloping Workplace Resilience and Managin g Stress\n\nThe first part of this session "strategies to build resilience " (1 hour) will provide a hands on discussion session working through the resilience toolkit. The second half will be a "Panel Discussion and Intera ctive session"— identifying your personal triggers, how to manage triggers and resilience.\n\n\nToni Collis (Appentra Solutions)\n------------------ ---\nUse Cases of Neuromorphic Co-Processors in Future HPC Environments\n\ nWith the looming end of Moore’s law and the end of Dennard scaling, the H PC community is exploring the use of specialized hardware as accelerators for certain tasks. Neuromorphic computing is a field in which neural netwo rks are implemented in hardware to achieve intelligent computation with lo wer p...\n\n\nCatherine Schuman (Oak Ridge National Laboratory)\n--------- ------------\nPower Aware Heterogeneous Node Assembly\n\nTo meet ever incr easing computational requirements, supercomputers and data centers are beg inning to utilize fat compute nodes with multiple hardware components such as manycore CPUs and accelerators. These components have intrinsic power variations even among same model components from same manufac...\n\n\nBilg e Acun (IBM Research)\n---------------------\nHigh Performance Computing i n Dynamic Traffic Simulation\n\nDynamic traffic simulation enables to emul ate the traffic congestion, which is a necessary aspect of traffic evoluti on. However, dynamic traffic simulations must compute in seconds if they are to be used in real-time traffic management systems. High performance c omputing (HPC) resources provide the...\n\n\nJuliette Ugirumurera (Lawrenc e Berkeley National Laboratory); Gabriel Gomes (University of California, Berkeley); Xiaoye Sherry Li (Lawrence Berkeley National Laboratory); and A lexandre Bayen (University of California, Berkeley)\n\nTag: Diversity, Edu cation, Hot Topics\n\nRegistration Category: Workshop Reg Pass\n\nSession Chairs: Toni Collis (Women in High Performance Computing); Weronika Filing er (Edinburgh Parallel Computing Centre (EPCC); University of Edinburgh, S cotland); and Misbah Mubarak (Amazon Web Services) END:VEVENT END:VCALENDAR