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:20200129T163600Z
LOCATION:607
DTSTART;TZID=America/Denver:20191118T155000
DTEND;TZID=America/Denver:20191118T161000
UID:submissions.supercomputing.org_SC19_sess124_ws_lasalss107@linklings.co
 m
SUMMARY:Generic Matrix Multiplication for Multi-GPU Accelerated Distribute
 d-Memory Platforms Over PaRSEC
DESCRIPTION:Workshop\n\nGeneric Matrix Multiplication for Multi-GPU Accele
 rated Distributed-Memory Platforms Over PaRSEC\n\nHerault, Robert, Bosilca
 , Dongarra\n\nThis paper introduces a generic and flexible matrix-matrix\n
   multiplication algorithm $C = A \times B$ for state-of-the-art\ncomput
 ing platforms. Typically, these platforms are\ndistributed-memory machin
 es whose nodes are equipped with several\naccelerators (e.g., 6 GPUs per
  node for Summit.  To\nthe best of our knowledge, SLATE is the only libr
 ary\nthat provides a publicly available implementation on such platforms
 ,\nand it is currently limited to problem instances where the $C$\nmat
 rix can entirely fit in the memory of the GPU accelerators.  Our\nalgori
 thm relies on the classical tile-based outer-product\nalgorithm, but enh
 ances it with several control dependences to\nincrease data re-use and t
 o optimize communication flow from/to the\naccelerators within each node
 . The algorithm is written within\nthe Parsec runtime system, which allo
 ws for a fast and generic\nimplementation, while achieving close-to-peak
  performance for a\nlarge variety of situations.\n\nTag: Workshop Reg Pa
 ss, Algorithms, Scalable Computing\n\nRegistration Category: Workshop Reg 
 Pass, Algorithms, Scalable Computing
URL:https://sc19.supercomputing.org/presentation/?id=ws_lasalss107&sess=se
 ss124
END:VEVENT
END:VCALENDAR