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:20211207T054803Z
LOCATION:230-231-232
DTSTART;TZID=America/Chicago:20211117T110000
DTEND;TZID=America/Chicago:20211117T113000
UID:submissions.supercomputing.org_SC21_sess168_pap468@linklings.com
SUMMARY:ElGA: Elastic and Scalable Dynamic Graph Analysis
DESCRIPTION:Paper\n\nElGA: Elastic and Scalable Dynamic Graph Analysis\n\n
 Gabert, Sancak, Ozkaya, Pinar, Catalyurek\n\nModern graphs are not only la
 rge, but rapidly changing.  The rate of change can vary significantly alon
 g with the computational cost.  Existing distributed graph analysis system
 s have largely been designed to operate on static graphs.  Infrastructure 
 changes in these systems need to occur when the system is idle, which can 
 result in significant wasted resources or the inability to cope with chang
 es.\n\nWe present ElGA, an elastic and scalable dynamic graph analysis sys
 tem.  Using a shared-nothing architecture and consistent hashing, ElGA can
  scale elastically as the graph grows or more computation is required.  By
  applying sketches, we perform an edge partitioning of the graph where hig
 h degree vertices can be split among multiple nodes.  ElGA supports both s
 ynchronous and asynchronous vertex-centric applications that operate in ba
 tches on a continuously changing graph.\n\nWe experimentally demonstrate t
 hat ElGA outperforms state-of-the-art static systems while supporting clie
 nt queries, elastic infrastructure changes, and dynamic algorithms.\n\nTag
 : Machine Learning and Artificial Intelligence, Numerical Algorithms\n\nRe
 gistration Category: Tech Program Reg Pass
END:VEVENT
END:VCALENDAR