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:20260422T000711Z
LOCATION:DEF Concourse
DTSTART;TZID=America/Denver:20231114T100000
DTEND;TZID=America/Denver:20231114T170000
UID:submissions.supercomputing.org_SC23_sess289_spostg115@linklings.com
SUMMARY:Fast Operations on Compressed Arrays without Decompression
DESCRIPTION:Harvey Dam (University of Utah)\n\nIn modern scientific comput
 ing and machine learning systems, data movement has overtaken compute as t
 he performance bottleneck, thus motivating the wider adoption of lossy dat
 a compression. Unfortunately, state-of-the-art floating-point array compre
 ssors such as SZ and ZFP require decompression before operations can be pe
 rformed on the data. In this work, our contribution is to show that compre
 ssion methods can be designed to allow efficient operations on compressed 
 arrays without having to first decompress. In particular, compression meth
 ods that consist of only linear transformations and quantization allow cer
 tain operations on compressed arrays without decompression. We develop suc
 h a compression method, called PyBlaz, the first compression method we kno
 w that can compress arbitrary-dimensional arrays and directly operate on t
 he compressed representation, with all stages running on GPUs.\n\nIn the p
 oster session, I will provide details about each compression step, several
  compressed-spaced operations, and our ongoing performance and application
  experiments.\n\nRegistration Category: Tech Program Reg Pass, Exhibits Re
 g Pass\n\n
END:VEVENT
END:VCALENDAR