BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20191206T142456Z
LOCATION:Analog 1\, 2
DTSTART;TZID=Europe/Stockholm:20180627T100000
DTEND;TZID=Europe/Stockholm:20180627T110000
UID:isc_hpc_ISC High Performance 2018_sess148_apost101@linklings.com
SUMMARY:(A01) Poster from China: A Low Communication Overhead Breadth-Firs
 t Search based on Global Bitmap .pdf
DESCRIPTION:HPC in Asia\n\n(A01) Poster from China: A Low Communication Ov
 erhead Breadth-First Search based on Global Bitmap .pdf\n\nDu, Peng, Cheng
 \n\nBreadth-First Search (BFS) is the underlying kernel algorithm for many
  graph applications such as social networks, medical informatics, transpor
 t systems, etc. Therefore, it has been absorbed as a core of Graph500, use
 d to evaluate the capability of supercomputers in terms of big data proces
 sing. In this paper, we introduce into a global bitmap which is used to ac
 celerate two approaches: the top-down and bottom-up. Specifically, the new
  top-down approach uses the global bitmap to indicate whether the vertices
  are visited or not, while the new bottom-up approach changes the frontier
  queue to the global bitmap to indicate whether the vertices are on the fr
 ontier. With the help of the global bitmap, the total number of communicat
 ion messages produced by the BFS will be reduced significantly, and conseq
 uentially the BFS is accelerated. Meanwhile, our algorithm is optimized fo
 r storage on Knights Landing (KNL). We evaluate our proposal on both the K
 NL platform and the Tianhe-2 supercomputer. Experimental results demonstra
 te that the communication was time reduced to roughly 1/4 of the original.
  We obtain speedups of 2.2-3.1 compared to the top-down approach.\n\nPasse
 s: Conference Pass, Clouds and Distributed Computing, Communication Optimi
 zation, Graph Algorithms, Parallel Algorithms\n\nTag: Conference Pass, Clo
 uds and Distributed Computing, Communication Optimization, Graph Algorithm
 s, Parallel Algorithms
URL:https://2018.isc-program.com/presentation/?id=apost101&sess=sess148
END:VEVENT
END:VCALENDAR