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:20190719T085744Z
LOCATION:HG EO Nord
DTSTART;TZID=Europe/Stockholm:20190613T195000
DTEND;TZID=Europe/Stockholm:20190613T215000
UID:submissions.pasc-conference.org_PASC19_sess179_post117@linklings.com
SUMMARY:CSM10 - Identifying Performance Challenges in Smoothed Particle Hy
 drodynamics Simulations
DESCRIPTION:Poster\n\n\nCSM10 - Identifying Performance Challenges in Smoo
 thed Particle Hydrodynamics Simulations\n\nMohammed, Cavelan, Ciorba, Cabe
 zon, Banicescu\n\nSmoothed particle hydrodynamics (SPH) is a commonly used
  technique for the numerical simulation of fluids in astrophysics and many
  computational science and engineering fields. SPH is grid-less, fully Lag
 rangian, particle method where the simulated system is discretized in a se
 ries of interpolation points (also called SPH-particles). The physical pro
 perties of the simulated astrophysical system are evaluated as a weighted 
 interpolation over close neighboring particles. SPH simulations are comput
 ationally-intensive and not trivial to parallelize. SPHYNX is a state-of-t
 he-art astrophysics SPH code that includes many recent upgrades, such as a
 daptive interpolating sinc kernels, an integral approach to calculate deri
 vatives, and generalized volume elements. The goal of this work is to stud
 y the performance of SPHYNX and identify its performance challenges on hig
 h performance computing (HPC) systems. To this end, certain performance is
 sues are identified, such as load imbalance at process and thread levels a
 nd non-overlapping communication and computation phases, among others. A n
 umber of improvements are proposed to address the identified challenges, a
 nd experiments on two HPC systems are performed to evaluate their effectiv
 eness. The performance results show improved application performance and b
 etter scaling profile.
END:VEVENT
END:VCALENDAR