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:20260522T150153Z
LOCATION:D220
DTSTART;TZID=America/Chicago:20181112T090000
DTEND;TZID=America/Chicago:20181112T173000
UID:submissions.supercomputing.org_SC18_sess172@linklings.com
SUMMARY:8th Workshop on Python for High-Performance and Scientific Computi
 ng
DESCRIPTION:Balsam: Automated Scheduling and Execution of Dynamic, Data-In
 tensive HPC Workflows\n\nWe introduce the Balsam service to manage high-th
 roughput task scheduling and execution on supercomputing systems. Balsam a
 llows users to populate a task database with a variety of tasks ranging fr
 om simple independent tasks to dynamic multi-task workflows. With abstract
 ions for the local resource s...\n\n\nMichael A. Salim, Thomas D. Uram, J.
  Taylor Childers, Venkatram Vishwanath, Michael E. Papka, and Prasanna Bal
 aprakash (Argonne National Laboratory)\n---------------------\nAcceleratin
 g the Signal Alignment Process in Time-Evolving Geometries Using Python\n\
 nThis paper addresses the computational challenges involved in postprocess
 ing of signals received using multiple collectors (satellites). Multiple l
 ow cost, small sized satellites can be used as dynamic beamforming arrays 
 (DBA) in remote sensing satellites. This usually requires precise metrolog
 y and...\n\n\nVinay B. Ramakrishnaiah and Zachary K. Baker (Los Alamos Nat
 ional Laboratory)\n---------------------\nWorkshop Morning Break\n--------
 -------------\nPanel: Interactivity in Supercomputing\n\nWilliam Scullin (
 Argonne National Laboratory) and Rollin Thomas (Lawrence Berkeley National
  Laboratory)\n---------------------\nIntroduction - 8th Workshop on Python
  for High-Performance and Scientific Computing\n\nPython is an established
 , high-level programming language with a large community in academia and i
 ndustry. Scientists, engineers, and educators use Python for data science,
  high-performance computing, and distributed computing. Since Python is ex
 tremely easy to learn with a very clean syntax, it is ...\n\n\nAndreas Sch
 reiber (German Aerospace Center), William Scullin (Argonne National Labora
 tory), Bill Spotz (Sandia National Laboratories), and Rollin Thomas (Lawre
 nce Berkeley National Laboratory)\n---------------------\nWorkshop Afterno
 on Break\n---------------------\nWorkshop Lunch (on your own)\n-----------
 ----------\nData-Parallel Python for High Energy Physics Analyses\n\nIn th
 is paper, we explore features available in Python which are useful for dat
 a reduction tasks in High Energy Physics (HEP). High-level abstractions in
  Python are convenient for implementing data reduction tasks. However, in 
 order for such abstractions to be practical, the efficiency of their perf.
 ..\n\n\nMarc Paterno, Christopher Green, Jim Kowalkowski, and Saba Sehrish
  (Fermi National Accelerator Laboratory)\n---------------------\nAutoParal
 lel: A Python Module for Automatic Parallelization and Distributed Executi
 on of Affine Loop Nests\n\nThe latest improvements in programming language
 s, programming models, and frameworks have focused on abstracting the user
 s from many programming issues. Among others, recent programming framework
 s include simpler syntax, automatic memory management and garbage collecti
 on, simplifies code re-usage th...\n\n\nCristian Ramon-Cortes, Ramon Amela
 , and Jorge Ejarque (Barcelona Supercomputing Center); Philippe Clauss (Fr
 ench Institute for Research in Computer Science and Automation (INRIA), Un
 iversity of Strasbourg); and Rosa M. Badia (Barcelona Supercomputing Cente
 r, Spanish National Research Council)\n---------------------\nManaging Pyt
 hon in HPC Environments\n\nPython has seen a rapid adoption in the weather
  and climate modeling science communities.  This swift rise has taken HPC 
 system administrators by surprise, leading to inadequate support.  These t
 rends, like those in other sciences, led to the development and widespread
  adoption of user managed binar...\n\n\nDaniel Gall (Engility Corporation)
  and Frank Indiviglio (National Oceanic and Atmospheric Administration)\n-
 --------------------\nPyHPC Lightning Talks\n\nWilliam Spotz (Sandia Natio
 nal Laboratories)\n---------------------\nKeynote: Better Scientific Softw
 are (BSSw)\n\nAnshu Dubey and Stephen Hudson (Argonne National Laboratory)
 \n---------------------\nPerformance, Power, and Scalability Analysis of t
 he Horovod Implementation of the CANDLE NT3 Benchmark on the Cray XC40 The
 ta\n\nTraining scientific deep learning models requires the large amount o
 f computing power provided by HPC systems. In this paper, we use the distr
 ibuted deep learning framework Horovod to parallelize NT3, a Python benchm
 ark from the exploratory research project CANDLE (Cancer Distributed Learn
 ing Enviro...\n\n\nXingfu Wu and Valerie Taylor (Argonne National Laborato
 ry, University of Chicago) and Justin M. Wozniak, Rick Stevens, Thomas Bre
 ttin, and Fangfang Xia (Argonne National Laboratory)\n\nTag: Parallel Appl
 ication Frameworks, Reproducibility, Scientific Computing\n\nRegistration 
 Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR