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:20181221T160909Z
LOCATION:C155/156
DTSTART;TZID=America/Chicago:20181114T133000
DTEND;TZID=America/Chicago:20181114T150000
UID:submissions.supercomputing.org_SC18_sess287_pan103@linklings.com
SUMMARY:Runtime for Exascale and Beyond: Convergence or Divergence?
DESCRIPTION:Panel\nTech Program Reg Pass\n\nRuntime for Exascale and Beyon
 d: Convergence or Divergence?\n\nSnir, Balaji, Kale, Sarkar, Treichler...\
 n\nWe propose to discuss in the panel the following three assertions:\n\na
 ) The coming decade is likely to see an accelerated evolution in programmi
 ng models for HPC, for several reasons:\n\n-- As Moore's law plateaus, in
 creased performance requires increasingly heterogeneous platforms and asyn
 chronous algorithms. \n\n-- As platforms become more dynamic (e.g., due t
 o dynamic power management) and algorithms become more adaptive, runtime a
 daptation is increasingly important.  Asynchronous Task Models (ATM), such
  as Charm++, Legion, PaRSEC, etc., seem better suited to handle this evolu
 tion than the prevalent model of MPI+OpenMP. \n\nb) The potential benefits
  of new programming models cannot fully realized without a more powerful r
 untime that better integrates internode and intranode synchronization and 
 communication with local resource scheduling and memory management.\n\nc) 
 The development and deployment of new programming models, as well as more 
 efficient support for current models can be accelerated by standardizing n
 ew interfaces that enable such integration
URL:https://sc18.supercomputing.org/presentation/?id=pan103&sess=sess287
END:VEVENT
END:VCALENDAR