BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241120T082410Z
LOCATION:HG E 1.2
DTSTART;TZID=Europe/Stockholm:20240604T150000
DTEND;TZID=Europe/Stockholm:20240604T153000
UID:submissions.pasc-conference.org_PASC24_sess181_pap113@linklings.com
SUMMARY:Using Read-After-Read Dependencies to Control Task-Granularity
DESCRIPTION:Paper\n\nAndres Gartmann (mynatix ag) and Mathias Müller (mete
 oblue ag)\n\nIn compiler theory, data analysis is used to exploit Instruct
 ion Level Parallelism (ILP). Three dependencies are used in modern compile
 rs and hardware schemes efficiently and are fundamental to any code compil
 ation. Read-after-read (RAR) has been left out, as it cannot cause a data 
 hazard. This article introduces a novel method to use the additional depen
 dence information contained in any code to enhance automatic parallelizati
 on. The method builds groups of arbitrary sequential instruction chains du
 ring static code analysis and introduces potential transfers between these
  groups. This gives new opportunities when optimizing code to a parallel p
 rocessing hardware. The segmentation enables more information concerning t
 he potential parallelization of the code and enhance optimization opportun
 ities to be gained during static code analysis. The novel principle is int
 roduced using a very simple example and then the segmentation is applied i
 n task- and data-parallelism examples. The automatic parallelization to a 
 multicore-platform is demonstrated based on the new segmentation method. T
 he ability to forecast the optimal distribution of the segments for a plat
 form with two key parameters and resulting codes are compared to measured 
 speedups.\n\nDomain: Computational Methods and Applied Mathematics\n\nSess
 ion Chair: Fazeleh Kazemian (Australian National University)
END:VEVENT
END:VCALENDAR