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:20260522T150116Z
LOCATION:D161
DTSTART;TZID=America/Chicago:20181112T115000
DTEND;TZID=America/Chicago:20181112T121000
UID:submissions.supercomputing.org_SC18_sess158_ws_lasalss105@linklings.co
 m
SUMMARY:Communication Avoiding Multigrid Preconditioned Conjugate Gradient
  Method for Extreme Scale Multiphase CFD Simulations
DESCRIPTION:Yasuhiro Idomura, Takuya Ina, Susumu Yamashita, Naoyuki Onoder
 a, and Susumu Yamada (Japan Atomic Energy Agency) and Toshiyuki Imamura (R
 IKEN)\n\nA communication avoiding (CA) multigrid preconditioned conjugate 
 gradient method  (CAMGCG) is applied to the pressure Poisson equation in a
  multiphase CFD code JUPITER, and its computational performance and conver
 gence property are compared against CA Krylov methods. A new geometric mul
 tigrid preconditioner is developed using a preconditioned Chebyshev iterat
 ion smoother, in which no global reduction communication is needed, halo d
 ata communication is reduced by a mixed precision approach, and eigenvalue
 s are computed using the CA Lanczos method. The CAMGCG solver has robust c
 onvergence properties regardless of the problem size, and shows both commu
 nication reduction and convergence improvement, leading to higher performa
 nce gain than CA Krylov solvers, which achieve only the former. The CAMGCG
  solver is applied to extreme scale multiphase CFD simulations with ~90 bi
 llion DOFs, and it is shown that compared with a preconditioned CG solver,
  the number of iterations, and thus, All\_Reduce is reduced to ~1/800, and
  ~11.6x speedup is achieved with keeping excellent strong scaling up to 8,
 000 KNLs on the Oakforest-PACS.\n\nTag: Algorithms, Heterogeneous Systems,
  Resiliency\n\nRegistration Category: Workshop Reg Pass\n\nSession Chairs:
  Vassil Alexandrov (Hartree Centre, STFC); Jack Dongarra (University of Te
 nnessee, Knoxville; Oak Ridge National Laboratory (ORNL)); Christian Engel
 mann (Oak Ridge National Laboratory (ORNL)); and Al Geist (Oak Ridge Natio
 nal Laboratory (ORNL))\n\n
END:VEVENT
END:VCALENDAR