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:20211207T055412Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211114T155500
DTEND;TZID=America/Chicago:20211114T160000
UID:submissions.supercomputing.org_SC21_sess328_ws_whpc106@linklings.com
SUMMARY:Early Career Lighting Talks - Balancing Particle GPU Memory Constr
 aints in Nyx Cosmology Code
DESCRIPTION:Workshop\n\nEarly Career Lighting Talks - Balancing Particle G
 PU Memory Constraints in Nyx Cosmology Code\n\nSexton\n\nThe Nyx cosmology
  code is used to simulate the formation of large scale structure in Lyman-
 alpha simulations of the universe. As the dark matter particles begin clus
 tering, the cost of a single time step grows due to the load imbalance. In
  particular, the Poisson solve for computing self-gravity becomes more exp
 ensive when there are highly clustered regions in the domain. These highly
  clustered regions can also prohibit fitting the entire problem in GPU HBM
 . If the entire problem does not fit on the HBM, Nyx should use managed me
 mory, and the cost of each time step becomes dependent on the host-device 
 memory bandwidth. In this talk, we will focus on the effect of load imbala
 nce on the GPU memory, as well as the current approach in Nyx to remedying
  this.\n\nTag: Online Only, Career Development, Diversity Equity Inclusion
  (DEI), Education and Training and Outreach, HPC Community Collaboration\n
 \nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR