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:20210916T132450Z
LOCATION:Ernesto Bertarelli
DTSTART;TZID=Europe/Stockholm:20210706T153000
DTEND;TZID=Europe/Stockholm:20210706T160000
UID:submissions.pasc-conference.org_PASC21_sess138_msa322@linklings.com
SUMMARY:In-Situ Visualization in Nek5000 with ParaView Catalyst
DESCRIPTION:Minisymposium\n\nIn-Situ Visualization in Nek5000 with ParaVie
 w Catalyst\n\nSchlatter\n\nIn-situ visualization on HPC systems allows us 
 to analyze simulation results that would otherwise be impossible, given th
 e size of the simulation data sets and offline post-processing execution t
 ime. We design and develop in-situ visualization with Paraview Catalyst in
  Nek5000, a massively parallel Fortran and C code for computational fluid 
 dynamics applications. We perform strong scalability tests up to 2048 core
 s on a Cray XC40 computer and assess in-situ visualization's impact on the
  Nek5000 performance. In our study case, a high-fidelity simulation of tur
 bulent flow, we observe that in-situ operations significantly limit the st
 rong scalability of the code, reducing the relative parallel efficiency to
  only 21% for frequent in-situ operations (the relative efficiency of Nek5
 000 without in-situ operations is about 99%). Through profiling with Arm M
 AP, we identified a bottleneck in the image composition step (that uses Ra
 dix-kr algorithm) where a majority of the time is spent on MPI communicati
 on. We also identified an imbalance of in-situ processing time between ran
 k 0 and all other ranks. Better scaling and load-balancing in the parallel
  image composition would considerably improve the performance and scalabil
 ity of Nek5000 with in-situ capabilities in large-scale simulation.\n\nDom
 ain: CS and Math, Physics, Engineering
END:VEVENT
END:VCALENDAR
