BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20220812T074335Z
LOCATION:Singapore Room
DTSTART;TZID=Europe/Stockholm:20220628T153000
DTEND;TZID=Europe/Stockholm:20220628T160000
UID:submissions.pasc-conference.org_PASC22_sess177_pap121@linklings.com
SUMMARY:FIST-HOSVD: Fused In-Place Sequentially Truncated Higher Order Sin
 gular Value Decomposition
DESCRIPTION:Paper\n\nFIST-HOSVD: Fused In-Place Sequentially Truncated Hig
 her Order Singular Value Decomposition\n\nCobb, Kolla, Phipps, Çatalyürek\
 n\nIn this paper, several novel methods of improving the memory locality o
 f the Sequentially Truncated Higher Order Singular Value Decomposition (ST
 -HOSVD) algorithm for computing the Tucker decomposition are presented. We
  show how the two primary computational kernels of the ST-HOSVD can be fus
 ed together into a single kernel to significantly improve memory locality.
  We then extend matrix tiling techniques to tensors to further improve cac
 he utilization. This block-based approach is then coupled with a novel in-
 place transpose algorithm to drastically reduce the memory requirements of
  the algorithm by overwriting the original tensor with the result. Our app
 roach's effectiveness is demonstrated by comparing the multi-threaded perf
 ormance of our optimized ST-HOSVD algorithm to TuckerMPI, a state-of-the-a
 rt ST-HOSVD implementation. We demonstrate up to ~135x reduction in memory
  consumption thereby increasing the problem size that can be computed for 
 a given memory allocation by up to ~3x, whilst maintaining comparable runt
 ime performance.\n\nDomain: Computer Science and Applied Mathematics
END:VEVENT
END:VCALENDAR