BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T143140Z
LOCATION:B304
DTSTART;TZID=America/New_York:20241118T114000
DTEND;TZID=America/New_York:20241118T115500
UID:submissions.supercomputing.org_SC24_sess749_ws_drbsd112@linklings.com
SUMMARY:Shifting Between Compute and Memory Bounds: A Compression-Enabled 
 Roofline Model
DESCRIPTION:Ramasoumya Naraparaju, Tianyu Zhao, Yanting Hu, and Dongfang Z
 hao (University of Washington) and Luanzheng Guo and Nathan Tallent (Pacif
 ic Northwest National Laboratory (PNNL))\n\nThis work proposes a compressi
 on-enabled roofline model to facilitate this adaptability with data compre
 ssion techniques to balance and transform between computational and memory
  demands. This model enables applications to adjust in response to the spe
 cific strengths and limitations of the underlying hardware and system to o
 ptimize resource utilization. The effectiveness of this approach is demons
 trated with matrix multiplication kernels on different input sizes, with t
 urning on/off various compression techniques, including 1) low-precision f
 loating point; 2) sparse matrix formulation; and 3) compressed arrays with
  ZFP. By reducing memory transfer volumes and cache misses and increasing 
 data locality and computational intensity through compression, the specifi
 c roofline model can transform between compute and memory bounds to align 
 more efficiently with system capabilities. This advancement not only impro
 ves overall performance but also maximizes adaptability in diverse computi
 ng environments.\n\nTag: Data Compression, Data Movement and Memory, Middl
 eware and System Software\n\nRegistration Category: Workshop Reg Pass\n\nS
 ession Chairs: Sheng Di (Argonne National Laboratory (ANL), University of 
 Chicago); Ana Gainaru (Oak Ridge National Laboratory (ORNL)); Sian Jin (Te
 mple University); Xin Liang (Oregon State University); Kento Sato (RIKEN C
 enter for Computational Science (R-CCS)); and Dingwen Tao (Institute of Co
 mputing Technology, Chinese Academy of Sciences; University of Chinese Aca
 demy of Sciences)\n\n
END:VEVENT
END:VCALENDAR