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:20260522T150110Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181114T083000
DTEND;TZID=America/Chicago:20181114T170000
UID:submissions.supercomputing.org_SC18_sess323_post249@linklings.com
SUMMARY:A Compiler Framework for Fixed-Topology Non-Deterministic Finite A
 utomata on SIMD Platforms
DESCRIPTION:Marziyeh Nourian, Hancheng Wu, and Michela Becchi (North Carol
 ina State University)\n\nAutomata traversal acceleration has been studied 
 on various parallel platforms. Many existing acceleration methods store fi
 nite automata states and transitions in memory. For these designs memory s
 ize and bandwidth are the main limiting factors to performance and power e
 fficiency. Many applications, however, require processing several fixed-to
 pology automata that differ only in the symbols associated to the transiti
 ons. This property enables the design of alternative, memory-efficient sol
 utions.  We target fixed-topology non-deterministic finite automata (NFAs)
  and propose a memory-efficient design, suitable to SIMD architectures, th
 at embeds the automata topology in code and stores only the transition sym
 bols in memory. We design a compiler that automates deployment of this des
 ign on SIMD platforms for a set of fixed-topology NFAs. Our compiler frame
 work performs a combination of platform-agnostic and platform-specific des
 ign decisions and optimizations. This poster describes the compiler toolch
 ain and shows the achieved throughput on GPU and Intel SIMD devices.\n\nRe
 gistration Category: Tech Program Reg Pass, Exhibits Reg Pass\n\n
END:VEVENT
END:VCALENDAR