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:20230124T171522Z
LOCATION:C147-154
DTSTART;TZID=America/Chicago:20221114T111400
DTEND;TZID=America/Chicago:20221114T113700
UID:submissions.supercomputing.org_SC22_sess452_ws_pawatm107@linklings.com
SUMMARY:Composition of Algorithmic Building Blocks in Template Task Graphs
DESCRIPTION:Workshop\n\nComposition of Algorithmic Building Blocks in Temp
 late Task Graphs\n\nHerault, Schuchart, Valeev, Bosilca\n\nIn this paper, 
 we explore the composition capabilities of the Template Task Graph (TTG) p
 rogramming model. We show how fine-grain composition of tasks is possible 
 in TTG between DAGs belonging to different libraries, even in a distribute
 d setup. We illustrate the benefits of this fine-grain composition on a li
 near algebra operation, the matrix inversion via the Cholesky method, whic
 h consists of three operations that need to be applied in sequence.\n\nEva
 luation on a cluster of many core shows that the transparent fine-grain co
 mposition implements the complex operation without introducing unnecessary
  synchronizations, increasing the overlap of communication and computation
 , and thus improving significantly the performance of the entire composed 
 operation.\n\nSession Format: Recorded\n\nTag: Applications, Architectures
 , Heterogeneous Systems, Hierarchical Parallelism, Parallel Programming La
 nguages and Models, Performance, Performance Portability, Scientific Compu
 ting\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR