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:20191206T142523Z
LOCATION:Panorama 3
DTSTART;TZID=Europe/Stockholm:20180627T110000
DTEND;TZID=Europe/Stockholm:20180627T123000
UID:isc_hpc_ISC High Performance 2018_sess166@linklings.com
SUMMARY:Challenges for Developing & Supporting HPC Applications
DESCRIPTION:Focus Session\n\nDeveloping and supporting HPC applications re
 quires more than just skill in using programming systems such as MPI, Open
 MP, and CUDA. Some of the issues that HPC users face every day include dat
 a management, performance and correctness debugging, integration with work
 flows, and limitations in the available programming systems and their impl
 ementations.  This session will describe current work to address some of t
 hese issues and provide a forum for the audience to ask questions about th
 e issues they face in developing and supporting HPC applications.\n\nParal
 lel Programming is the Easy Part: The Real Challenges in Developing HPC Ap
 plications\n\nGropp\n\nIn discussing the software crisis in high performan
 ce computing, the focus is often on the issues in writing and debugging pa
 rallel programs.  In fact, the claim that there is a software crisis is so
 metimes based on the difficulty in constructing high performance, scalable
  applications.  But there a...\n\n---------------------\nModeling the Next
 -Generation High Performance Schedulers\n\nTaufer\n\nHigh performance comp
 uting (HPC) resources and workloads are undergoing tumultuous changes. HPC
  resources are growing more diverse with the adoption of accelerators; HPC
  workloads have increased in size by orders of magnitude. Despite these ch
 anges, when assigning workload jobs to resources, HPC sch...\n\n----------
 -----------\nBugs and Speed in HPC Applications: Past, Present, and Future
 \n\nHollingsworth\n\nThere is a bug in my program. My application doesn’t 
 run fast enough.  These have been two of the most enduring and ever presen
 t attributes of HPC, probably computing in general.  For almost as long, r
 esearchers have endeavored to create tools to help solve those problems.  
 In this session, I will b...\n\n---------------------\nAutomatic compiler-
 driven GPU acceleration with Polly-ACC\n\nHoefler\n\nProgramming today's i
 ncreasingly complex heterogeneous hardware is difficult, as it commonly re
 quires the use of data-parallel languages, pragma annotations, specialized
  libraries, or DSL compilers. Adding explicit accelerator support into a l
 arger code base is not only costly, but also introduces a...\n\n\nPasses: 
 Conference Pass
END:VEVENT
END:VCALENDAR