BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20211207T055412Z
LOCATION:222
DTSTART;TZID=America/Chicago:20211118T153000
DTEND;TZID=America/Chicago:20211118T173000
UID:submissions.supercomputing.org_SC21_sess454@linklings.com
SUMMARY:Doctoral Showcase III Presentations
DESCRIPTION:Doctoral Showcase, Posters\n\nMemory-Centric 3D Image Reconstr
 uction with Hierarchical Communications on Multi-GPU Node Architecture\n\n
 Hidayetoglu, Hwu\n\nX-ray computed tomography is a commonly used technique
  for noninvasive imaging at synchrotron facilities. Iterative tomographic 
 reconstruction algorithms are often preferred for recovering high quality 
 3D volumetric images from 2D X-ray images, however, their use has been lim
 ited to small/medium dat...\n\n---------------------\nPerformance Profilin
 g, Analysis, and Optimization of GPU-Accelerated Applications\n\nZhou, Mel
 lor-Crummey\n\nGPUs have emerged as a key component for accelerating appli
 cations in various domains, including deep learning, data analytics, and s
 cientific simulations. While GPUs provide superior compute power and highe
 r memory bandwidth than CPUs, writing efficient GPU code to achieve maximu
 m possible performa...\n\n---------------------\nHolistic Performance Anal
 ysis and Optimization of Unified Virtual Memory\n\nAllen, Ge\n\nHigh-perfo
 rmance computing systems have seen tremendous growth in theoretical perfor
 mance with the inclusion of Graphics Processing Units (GPUs) and other acc
 elerators. The difficulty of programming these systems has grown alongside
  the performance as programmers are required to manage separate prog...\n\
 n---------------------\nAccelerating I/O for Traditional HPC and Modern ML
  Workloads on Emerging HPC Systems\n\nChien, Markidis, Podobas\n\nIn recen
 t years, HPC systems have emerged as an attractive option to speed up larg
 e-scale Machine Learning (ML) workloads. HPC systems can tremendously impr
 ove learning speed with lots of GPUs and fast interconnect. However, ML wo
 rkloads that are data-intensive differ significantly from traditional ...\
 n\n\nRegistration Category: Tech Program Reg Pass
END:VEVENT
END:VCALENDAR