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:20181221T160904Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181113T083000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess325_spost115@linklings.com
SUMMARY:OoO Instruction Benchmarking Framework on the Back of Dragons
DESCRIPTION:ACM Student Research Competition, Poster\nTech Program Reg Pas
 s, Exhibits Reg Pass\n\nOoO Instruction Benchmarking Framework on the Back
  of Dragons\n\nHammer\n\nIn order to construct an accurate instruction exe
 cution model for modern out-of-order micro architectures, an accurate desc
 ription of instruction latency, throughput and concurrency is indispensabl
 e. Already existing resources and vendor provided information is neither c
 omplete nor detailed enough and sometimes incorrect. We therefore proclaim
  to deduct this information through runtime instruction benchmarking and p
 resent a framework to support such investigations based on LLVM's just-in-
 time and cross-platform compilation capabilities.\n\npyasmjit abstracts in
 structions, operands, and dependency chains, to easily construct the neces
 sary benchmarks. The synthesized code is interactively compiled and execut
 ed using the llvmlite library, which in turn is based on the stable LLVM C
 -API.  pyasmjit offers a command line as well as a programming interface. 
  Unlike other approaches, we do not rely on model specific performance cou
 nters and focus on interoperability and automation to support quick modeli
 ng of future microarchitectures.
URL:https://sc18.supercomputing.org/presentation/?id=spost115&sess=sess325
END:VEVENT
END:VCALENDAR