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:20190719T085745Z
LOCATION:HG F 3
DTSTART;TZID=Europe/Stockholm:20190614T110000
DTEND;TZID=Europe/Stockholm:20190614T113000
UID:submissions.pasc-conference.org_PASC19_sess123_msa113@linklings.com
SUMMARY:Phase Asynchronous AMR Execution on Heterogeneous Architectures
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Physi
 cs, Solid Earth Dynamics\n\nPhase Asynchronous AMR Execution on Heterogene
 ous Architectures\n\nUnat\n\nIn the Adaptive Mesh Refinement (AMR) context
 , porting an application from synchronous to fully asynchronous execution 
 increases application performance but also requires tremendous programming
  effort because the programmer has to explicitly specify data dependencies
 , schedule a grid for computation upon its data dependencies met and issue
  data communication requests upon computation of a grid. Previously at our
  Supercomputing 2018 work, we showed that phase asynchronous execution for
  AMR provides an application with both the performance and productivity be
 nefits. In this work, we extend the phase asynchrony to GPUs and propose s
 olutions to the challenges that arise while porting AMR application on het
 erogeneous architectures.
END:VEVENT
END:VCALENDAR