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:20260522T150111Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181115T083000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess324_post108@linklings.com
SUMMARY:Kernel-Based and Total Performance Analysis of CGYRO on 4 Leadersh
 ip Systems
DESCRIPTION:Igor Sfiligoi, Jeff Candy, and Emily Belli (General Atomics)\n
 \nWe present the results of an exhaustive performance analysis of the CGYR
 O code on 4 leadership systems spanning 5 different configurations (2 KNL-
 based, 1 Skylake-based, and 2 hybrid CPU-GPU architectures). CGYRO is an E
 ulerian gyrokinetic solver designed and optimized for collisional, electro
 magnetic, multiscale fusion plasma simulation. It is based on the well-kno
 wn GYRO code, but redesigned from the ground up to operate efficiently on 
 multicore and GPU-accelerated systems. The gyrokinetic equations specify a
  5-dimensional distribution function for each species, with species couple
 d through both the Maxwell equations and collision operator. For the cross
 -machine performance analysis, we report and compare timings for 4 computa
 tional and 4 communication kernels. This kernel-based breakdown illustrate
 s the strengths and weaknesses of the floating-point and communication arc
 hitectures of the respective systems. An overview of the physical equation
 s solved, the scalable numerical methods used, and data communication patt
 erns required by each kernel are also given.\n\nRegistration Category: Tec
 h Program Reg Pass, Exhibits Reg Pass\n\n
END:VEVENT
END:VCALENDAR