BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240626T180033Z
LOCATION:Level 2 Lobby
DTSTART;TZID=America/Los_Angeles:20240625T180000
DTEND;TZID=America/Los_Angeles:20240625T190000
UID:dac_DAC 2024_sess236_RESEARCH968@linklings.com
SUMMARY:Algorithm-Hardware Co-Design of Distribution-Aware Logarithmic-Pos
 it Encodings for Efficient DNN Inference
DESCRIPTION:Work-in-Progress Poster\n\nHesham Mostafa and Adam Grabowski (
 Intel Corporation); Md Asadullah Turja (University of North Carolina, Chap
 el Hill); Juan Cervino and Alejandro Ribeiro (University of Pennsylvania);
  and Nageen Himayat (Intel Corporation)\n\nTraining Graph Neural Networks(
 GNNs) on a large monolithic graph presents unique challenges as the graph 
 cannot fit within a single machine and it cannot be decomposed into smalle
 r disconnected components. Distributed sampling-based training distributes
  the graph across multiple machines and trains the GNN on small parts of t
 he graph that are randomly sampled every training iteration. We show that 
 in a distributed environment, the sampling overhead is a significant compo
 nent of the training time for large-scale graphs. We propose FastSample wh
 ich is composed of two synergistic techniques that greatly reduce the dist
 ributed sampling time: 1)~a new graph partitioning method that eliminates 
 most of the communication rounds in distributed sampling , 2)~a novel high
 ly optimized sampling kernel that reduces memory movement during sampling.
  We test FastSample on large-scale graph benchmarks and show that FastSamp
 le speeds up distributed sampling-based GNN training by up to 2x with no l
 oss in accuracy.\n\nTopic: AI, Autonomous Systems, Cloud, Design, EDA, Emb
 edded Systems, IP, Security
END:VEVENT
END:VCALENDAR