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:20211207T055401Z
LOCATION:224
DTSTART;TZID=America/Chicago:20211115T120000
DTEND;TZID=America/Chicago:20211115T123000
UID:submissions.supercomputing.org_SC21_sess350_ws_ia108@linklings.com
SUMMARY:No More Leaky PageRank
DESCRIPTION:Workshop\n\nNo More Leaky PageRank\n\nSallinen, Ripeanu\n\nWe 
 have surveyed multiple PageRank implementations available with popular gra
 ph processing frameworks, and discovered that they treat sink vertices (i.
 e., vertices without outgoing edges) incorrectly. This leads to two issues
 : (i) incorrect PageRank scores, and (ii) flawed performance evaluations (
 as costly scatter operations are avoided). For synchronous PageRank implem
 entations, a strategy to fix these issues exists (accumulating values from
  sinks during an algorithmic superstep of a PageRank iteration), albeit wi
 th sizeable overhead. This solution, however, is not applicable to asynchr
 onous frameworks.\n\nWe present and evaluate a novel, low-cost algorithmic
  solution to address this issue. For asynchronous PageRank, our solution s
 imply requires an inexpensive O(Vertex) computation performed alongside th
 e final normalization step. We also show that this strategy has advantages
  over prior work for synchronous PageRank, as it both avoids graph restruc
 turing and reduces inline computation costs by performing a final score re
 assignment to vertices once at the end of processing.\n\nTag: Algorithms, 
 Architectures, Big Data, Data Analytics, Memory Systems, Numerical Algorit
 hms\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR