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:20181221T160725Z
LOCATION:D175
DTSTART;TZID=America/Chicago:20181111T090000
DTEND;TZID=America/Chicago:20181111T090500
UID:submissions.supercomputing.org_SC18_sess155_wksp120@linklings.com
SUMMARY:Introduction - Fifth Workshop on Accelerator Programming Using Dir
 ectives (WACCPD)
DESCRIPTION:Workshop\nAccelerators, Heterogeneous Systems, Parallel Progra
 mming Languages, Libraries, and Models, Workshop Reg Pass\n\nIntroduction 
 - Fifth Workshop on Accelerator Programming Using Directives (WACCPD)\n\nJ
 uckeland, Chandrasekaran, Wienke\n\nCurrent hardware trends lead to ever m
 ore complex compute node architectures offering multiple, heterogeneous le
 vels of massive parallelism. As a result, the ‘X’ in MPI+X demands more fo
 cus. In order to exploit the maximum available parallelism out of such sys
 tems, we are in dire need of sophisticated programming approaches that can
  provide scalable as well as portable solutions without compromising on pe
 rformance. The expectation from the scientific community is that such solu
 tions should allow programmers to maintain a single code base whenever pos
 sible and to avoid requiring maintaining and debug multiple versions of th
 e same code.  \n\nRaising the abstraction of the code is one of the effect
 ive methodologies to reduce the burden on the programmer. At the same time
  such a methodology will require a heroic compiler to be designed. Softwar
 e abstraction-based programming models such as OpenMP and OpenACC have bee
 n serving this purpose over the past several years. These programming mode
 ls address the ‘X’ component by providing programmers high-level directive
 -based approaches to accelerate and port scientific applications to hetero
 geneous platforms. \n\nThe focus of this workshop is to explore this ‘X’ c
 omponent in a hybrid MPI+X programming approach. We present technical pape
 rs discussing innovative high-level language features and their (early pro
 totype) implementations needed to address hierarchical heterogeneous syste
 ms, stories and lessons learned while using directives to migrate scientif
 ic legacy code to parallel processors, state-of-the-art compilation and ru
 ntime scheduling techniques, techniques to optimize performance, mechanism
 s to keep communication and synchronization efficient.
URL:https://sc18.supercomputing.org/presentation/?id=wksp120&sess=sess155
END:VEVENT
END:VCALENDAR