BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20200129T163558Z
LOCATION:704-706
DTSTART;TZID=America/Denver:20191117T113000
DTEND;TZID=America/Denver:20191117T120000
UID:submissions.supercomputing.org_SC19_sess114_ws_prot108@linklings.com
SUMMARY:Designing Efficient Parallel Software via Compositional Performanc
 e Modeling
DESCRIPTION:Workshop\n\nDesigning Efficient Parallel Software via Composit
 ional Performance Modeling\n\nCalotoiu, Höhl, Mantel, Nguyen, Wolf\n\nPerf
 ormance models are powerful instruments for understanding the performance 
 of parallel systems and uncovering their bottlenecks. Already during syste
 m design, performance models can help ponder alternatives. However, creati
 ng a performance model - whether theoretically or empirically - for an ent
 ire application that does not exist yet is challenging unless the interact
 ions between all system components are well understood, which is often not
  the case during design.  In this paper, we propose to generate performanc
 e models of full programs from performance models of their components usin
 g formal composition operators derived from parallel design patterns such 
 as pipeline or task pool. As long as the design of the overall system foll
 ows such a pattern, its performance model can be predicted with reasonable
  accuracy without an actual implementation.\n\nTag: Workshop Reg Pass, Per
 formance, Programming Systems, Visualization\n\nRegistration Category: Wor
 kshop Reg Pass, Performance, Programming Systems, Visualization
URL:https://sc19.supercomputing.org/presentation/?id=ws_prot108&sess=sess1
 14
END:VEVENT
END:VCALENDAR