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:20260522T150119Z
LOCATION:D168
DTSTART;TZID=America/Chicago:20181116T093500
DTEND;TZID=America/Chicago:20181116T095000
UID:submissions.supercomputing.org_SC18_sess144_ws_pawatm101@linklings.com
SUMMARY:Comparison of the HPC and Big Data Java Libraries Spark, PCJ and A
 PGAS
DESCRIPTION:Jonas Posner, Claudia Fohry, and Lukas Reitz (University of Ka
 ssel)\n\nAlthough Java is rarely used in HPC, there are a few notable libr
 aries. Use of Java may help to bridge the gap between HPC and big data pro
 cessing.\n\nThis paper compares the big data library Spark, and the HPC li
 braries PCJ and APGAS, regarding productivity and performance.  We refer t
 o Java versions for all libraries. For APGAS, we include both the original
  version and an own extension by locality-flexible tasks. We consider thre
 e benchmarks: Calculation of pi from HPC, Unbalanced Tree Search (UTS) fro
 m HPC, and WordCount from the big data domain.\n\nIn performance measureme
 nts with up to 144~workers, the extended APGAS library was the clear winne
 r. With 144 workers, APGAS programs were up to a factor of more than two f
 aster than Spark programs, and up to about 30% faster than PCJ programs. R
 egarding productivity, the extended APGAS programs consistently needed the
  lowest number of different library constructs. Spark ranged second in pro
 ductivity, and PCJ third.\n\nTag: Parallel Programming Languages, Librarie
 s, and Models, Productivity\n\nRegistration Category: Workshop Reg Pass\n\
 n
END:VEVENT
END:VCALENDAR