BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160559Z
LOCATION:Track 8
DTSTART;TZID=America/New_York:20201113T170000
DTEND;TZID=America/New_York:20201113T170500
UID:submissions.supercomputing.org_SC20_sess227_pec319@linklings.com
SUMMARY:DaCe Python Frontend
DESCRIPTION:Workshop\n\nDaCe Python Frontend\n\nZiogas\n\nSeveral framewor
 ks for developing HPC applications have identified the high cost of data m
 ovement as the primary performance bottleneck in modern computing architec
 tures. The Data-Centric (DaCe) Parallel Programming framework uses the Sta
 teful DataFlow multiGraph (SDFG) intermediate representation, which provid
 es a transformable and interactive view of the input code. This data-centr
 ic view exposes to the user/performance engineer optimization opportunitie
 s that go beyond the boundaries of low-level computational kernels and enc
 ompass the whole program.\n\nDaCe users can create SDFGs by either using a
 n API or writing code in one of the supported languages, including Python.
  Therefore, scientists and engineers can convert their Python codebase to 
 SDFGs with relative ease and take advantage of the optimization capabiliti
 es of the DaCe framework. In this talk, we would like to present this fron
 tend, which supports a subset of Python and the NumPy module and discuss t
 he challenges in converting a functional representation to dataflow. We al
 so want to highlight issues that arise due to language semantics and progr
 ammer expectations, such as handling differences in data types, conversion
 s and operator semantics. Moreover, we want to showcase the framework’s ca
 pabilities in integrating with HPC libraries, e.g., BLAS.\n\nRegistration 
 Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR