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:20260522T150123Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181113T083000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess322_post247@linklings.com
SUMMARY:Compiling SIMT Programs on Multi- and Many-Core Processors with Wi
 de Vector Units: A Case Study with CUDA
DESCRIPTION:Hancheng Wu, John Ravi, and Michela Becchi (North Carolina Sta
 te University)\n\nThere has been an increasing interest in SIMT programmin
 g tools for multi- and manycore (co)processors with wide vector extensions
 . In this work, we study the effective implementation of a SIMT programmin
 g model (a subset of CUDA C) on Intel platforms with 512-bit vector extens
 ions (hybrid MIMD/SIMD architectures). We first propose a set of compiler 
 techniques to transform programs written using a SIMT programming model in
 to code that leverages both the x86 cores and the vector units of a hybrid
  MIMD/SIMD architecture, thus providing programmability, high system utili
 zation and portability. We then evaluate the proposed techniques on variou
 s hybrid systems using microbenchmarks and real-world applications. Finall
 y, we point out the main challenges in supporting the SIMT model on hybrid
  systems.\n\nRegistration Category: Tech Program Reg Pass, Exhibits Reg Pa
 ss\n\n
END:VEVENT
END:VCALENDAR