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:20260522T150111Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181115T083000
DTEND;TZID=America/Chicago:20181115T170000
UID:submissions.supercomputing.org_SC18_sess324_post152@linklings.com
SUMMARY:Applying the Execution-Cache-Memory Model: Current State of Practi
 ce
DESCRIPTION:Georg Hager, Jan Eitzinger, and Julian Hornich (University of 
 Erlangen-Nuremberg, Erlangen Regional Computing Center); Francesco Cremone
 si (Swiss Federal Institute of Technology in Lausanne); Christie L. Alappa
 t (University of Erlangen-Nuremberg, Erlangen Regional Computing Center); 
 Thoams Roehl (University of Erlangen-Nuremberg); and Gerhard Wellein (Univ
 ersity of Erlangen-Nuremberg, Erlangen Regional Computing Center)\n\nThe E
 CM (Execution-Cache-Memory) model is an analytic, resource-based  performa
 nce model for steady-state loop code running on multicore processors. Star
 ting from a machine model, which describes the interaction between the cod
 e and the hardware, and static code analysis, it allows an accurate predic
 tion of the runtime of sequential loop code. Together with a scaling assum
 ption, it also gives a performance scaling prediction. This poster summari
 zes the current state of practice in constructing and applying the ECM mod
 el, points out problems and open questions, and applies the model to three
  new and nontrivial use cases. For the first time, overlap assumptions for
  all relevant CPU architectures in high performance computing are presente
 d.\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg Pass\n\n
END:VEVENT
END:VCALENDAR