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:20181221T160731Z
LOCATION:C140/142
DTSTART;TZID=America/Chicago:20181115T153000
DTEND;TZID=America/Chicago:20181115T160000
UID:submissions.supercomputing.org_SC18_sess219_pap239@linklings.com
SUMMARY:Phase Asynchronous AMR Execution for Productive and Performant Ast
 rophysical Flows
DESCRIPTION:Paper\nAlgorithms, Applications, Computational Physics, Scient
 ific Computing, Tech Program Reg Pass\n\nPhase Asynchronous AMR Execution 
 for Productive and Performant Astrophysical Flows\n\nFarooqi, Nguyen, Zhan
 g, Almgren, Shalf...\n\nAdaptive Mesh Refinement (AMR) is an approach to s
 olving PDEs that reduces the computational and memory requirements at the 
 expense of increased communication. Although adopting asynchronous executi
 on can overcome communication issues, manually restructuring an AMR applic
 ation to realize asynchrony is extremely complicated and hinders readabili
 ty and long-term maintainability. To balance performance against productiv
 ity, we design a user-friendly API and adopt phase asynchronous execution 
 model where all subgrids at an AMR level can be computed asynchronously. \
 n\nWe apply the phase asynchrony to transform a real-world AMR application
 , CASTRO, which solves multicomponent compressible hydrodynamic equations 
 for astrophysical flows. We evaluate the performance and programming effor
 t required to use our carefully designed API and execution model for trans
 itioning large legacy codes from synchronous to asynchronous execution up 
 to 278,528 Intel-KNL cores. CASTRO is about 100K lines of code but less th
 an 0.2% code changes are required to achieve significant performance impro
 vement.
URL:https://sc18.supercomputing.org/presentation/?id=pap239&sess=sess219
END:VEVENT
END:VCALENDAR

