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:20240626T180034Z
LOCATION:3008\, 3rd Floor
DTSTART;TZID=America/Los_Angeles:20240627T151500
DTEND;TZID=America/Los_Angeles:20240627T153000
UID:dac_DAC 2024_sess150_RESEARCH1457@linklings.com
SUMMARY:A Combined Content Addressable Memory and In-Memory Processing App
 roach for k-Clique Counting Acceleration
DESCRIPTION:Research Manuscript\n\nXidi Ma, Weichen Zhang, and Xueyan Wang
  (Beihang University); Tianyang Yu and Bi Wu (Nanjing University of Aerona
 utics and Astronautics); Gang Qu (Univ. of Maryland, College Park); and We
 isheng Zhao (Beihang University)\n\nk-Clique counting problem plays an imp
 ortant role in graph mining which has seen a growing number of application
 s. However, current k-Clique counting accelerators cannot meet the perform
 ance requirement mainly because they struggle with high data transfer issu
 e incurred by the intensive set intersection operations and the inability 
 of load balancing. In this paper, we propose to solve this problem with a 
 hybrid framework of content addressable memory (CAM) and in-memory process
 ing (PIM). Specifically, we first utilize CAM for binary induced subgraph 
 generation in order to reduce the search space, then we use PIM to impleme
 nt in-place parallel k-Clique counting through iterative Boolean logic "AN
 D"- like operation. To take full advantage of this combined CAM and PIM fr
 amework, we develop dynamic task scheduling strategies that can achieve ne
 ar optimal load balancing among the PIM arrays. Experimental results demon
 strate that, compared with state-of-the-art CPU and GPU platforms, our app
 roach achieves speedups of 167.5× and 28.8×, respectively. Meanwhile, the 
 energy efficiency is improved by 788.3× over the GPU baseline.\n\nTopic: E
 mbedded Systems\n\nKeyword: Embedded Memory and Storage Systems\n\nSession
  Chair: Filippo Carloni (Politecnico di Milano)
END:VEVENT
END:VCALENDAR