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:20190719T085744Z
LOCATION:HG D 1.1
DTSTART;TZID=Europe/Stockholm:20190613T111500
DTEND;TZID=Europe/Stockholm:20190613T114500
UID:submissions.pasc-conference.org_PASC19_sess125_msa172@linklings.com
SUMMARY:Towards Fault-Tolerance in Atmospheric Dynamical Cores
DESCRIPTION:Minisymposium\nComputer Science and Applied Mathematics, Clima
 te and Weather\n\nTowards Fault-Tolerance in Atmospheric Dynamical Cores\n
 \nBenacchio, Bonaventura\n\nNumerical weather prediction and climate studi
 es rate among the most compute-intensive applications in applied mathemati
 cs. Spatial resolutions for atmospheric models currently in the sub-10 km 
 scale force the use of massive core counts, which in turn entail a rapidly
  growing likelihood of hard and soft faults, errors, and failures. In this
  landscape, standard checkpointing strategies currently used in research a
 nd operational codes look increasingly unsustainable and will soon fall sh
 ort of the required production schedules. This talk will introduce the the
 mes in the minisymposium with a survey of available approaches for fault-t
 olerance and data recovery, focusing on numerical methods for atmospheric 
 dynamics. Most time integration schemes in this field feature implicitness
  in order to sidestep stability constraints on the time step size, thereby
  requiring the construction and solution of large linear systems. Strategi
 es to endow existing and future linear solvers with resiliency against har
 d and soft faults will be discussed, with particular attention to their im
 pact on efficiency and workflows.
END:VEVENT
END:VCALENDAR