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:20230124T171521Z
LOCATION:C1-2-3
DTSTART;TZID=America/Chicago:20221117T083000
DTEND;TZID=America/Chicago:20221117T170000
UID:submissions.supercomputing.org_SC22_sess275_rpost155@linklings.com
SUMMARY:Custom 8-bit Floating Point Value Format for Reducing Shared Memor
 y Bank Conflict in Approximate Nearest Neighbor Search
DESCRIPTION:Posters, Research Posters\n\nCustom 8-bit Floating Point Value
  Format for Reducing Shared Memory Bank Conflict in Approximate Nearest Ne
 ighbor Search\n\nOotomo, Naruse\n\nThe k-nearest neighbor search is used i
 n various applications such as machine learning, computer vision, database
  search, and information retrieval. While the computational cost of the ex
 act nearest neighbor search is enormous, an approximate nearest neighbor s
 earch (ANNS) is being paid much attention. IVFPQ is one of the ANNS method
 s. Although we can leverage the high bandwidth and low latency of shared m
 emory to compute the search phase of the IVFPQ on NVIDIA GPUs, the through
 put can degrade due to shared memory bank conflict. To reduce the bank con
 flict and improve the search throughput, we propose a custom 8-bit floatin
 g point value format. This format doesn’t have a sign bit and can be conve
 rted from/to FP32 with a few instructions. We use this format for IVFPQ on
  GPUs and get better performance without significant recall loss compared 
 to FP32 and FP16.\n\nRegistration Category: Tech Program Reg Pass, Exhibit
 s Reg Pass
END:VEVENT
END:VCALENDAR