BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000712Z
LOCATION:702
DTSTART;TZID=America/Denver:20231112T163000
DTEND;TZID=America/Denver:20231112T170000
UID:submissions.supercomputing.org_SC23_sess436_ws_ia103@linklings.com
SUMMARY:cuAlign: Scalable Network Alignment on GPU Accelerators
DESCRIPTION:Lizhi Xiang (University of Utah), Arif Khan (Meta), S. M. Ferd
 ous (Pacific Northwest National Laboratory (PNNL)), SR Aravind (Meta), and
  Mahantesh Halappanavar (Pacific Northwest National Laboratory (PNNL))\n\n
 Given two graphs, the objective of network alignment is to find a one-to-o
 ne mapping of vertices in one graph to vertices in the other, such that th
 e number of overlaps is maximized. Network alignment is an important optim
 ization problem with several applications in bioinformatics, computer visi
 on and ontology matching. Since it is an NP-hard problem, efficient heuris
 tics and scalable implementations are necessary. In this work, we introduc
 e a novel framework (cuAlign) that combines intra-network proximity using 
 node (vertex) embedding, sparsification for computational efficiency, and 
 belief propagation (BP) and approximate weighted matching for alignment. W
 e demonstrate qualitative improvements up to 22% over state-of-the-art app
 roaches and provide a scalable implementation targeting modern GPU acceler
 ators. We demonstrate up to 19× speedup for belief propagation, 3× speedup
  for approximate weighted matching, and 15× total, relative to a state-of-
 the-art multi-threaded implementation. We believe that our work will enabl
 e algorithmic improvements and applications of network alignment.\n\nTag: 
 Algorithms, Applications, Architecture and Networks\n\nRegistration Catego
 ry: Workshop Reg Pass\n\nSession Chairs: Sanjukta Bhowmick (University of 
 North Texas), Vito Giovanni Castellana (Pacific Northwest National Laborat
 ory (PNNL)), John Feo (Pacific Northwest National Laboratory (PNNL)), Marc
 o Minutoli (Advanced Micro Devices), and Antonino Tumeo (Pacific Northwest
  National Laboratory (PNNL))\n\n
END:VEVENT
END:VCALENDAR