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:20210402T160106Z
LOCATION:Track 3
DTSTART;TZID=America/New_York:20201119T160000
DTEND;TZID=America/New_York:20201119T163000
UID:submissions.supercomputing.org_SC20_sess160_pap315@linklings.com
SUMMARY:Rocket: Efficient and Scalable All-Pairs Computations on Heterogen
 eous Platforms
DESCRIPTION:Paper\n\nRocket: Efficient and Scalable All-Pairs Computations
  on Heterogeneous Platforms\n\nHeldens, Hijma, van Werkhoven, Maassen, Bal
 ...\n\nAll-pairs compute problems apply a user-defined function to each co
 mbination of two items of a given data set. Although these problems presen
 t an abundance of parallelism, data reuse must be exploited to achieve goo
 d performance. Several researchers considered this problem, either resorti
 ng to partial replication with static work distribution or dynamic schedul
 ing with full replication. In contrast, we present a solution that relies 
 on hierarchical multi-level software-based caches to maximize data reuse a
 t each level in the distributed memory hierarchy combined with a divide-an
 d-conquer approach to exploit data locality, hierarchical work-stealing to
  dynamically balance the workload and asynchronous processing to maximize 
 resource utilization. We evaluate our solution using three real-world appl
 ications, from digital forensics, localization microscopy and bioinformati
 cs, on different platforms, from desktop machine to a supercomputer. Resul
 ts shows excellent efficiency and scalability when scaling to 96 GPUs, eve
 n obtaining super-linear speedups due to a distributed cache.\n\nTag: Grap
 h Algorithms, Scalable Computing\n\nRegistration Category: Tech Program Re
 g Pass
END:VEVENT
END:VCALENDAR

