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:20260422T000609Z
LOCATION:505
DTSTART;TZID=America/Denver:20231115T103000
DTEND;TZID=America/Denver:20231115T120000
UID:submissions.supercomputing.org_SC23_sess307@linklings.com
SUMMARY:Best Research Poster Presentations
DESCRIPTION:Improving Memory Interfacing in HLS-Generated Accelerators wit
 h Custom Caches\n\nAccelerators based on reconfigurable devices are becomi
 ng popular for data analytics in high performance computing and cloud comp
 uting systems. However, designing these accelerators is a hard problem. Hi
 gh-Level Synthesis tools can help by generating RTL designs from high-leve
 l languages, but they t...\n\n\nClaudio Barone (Pacific Northwest National
  Laboratory (PNNL)), Giovanni Gozzi and Michele Fiorito (Polytechnic Unive
 rsity of Milan), Ankur Limaye and Antonino Tumeo (Pacific Northwest Nation
 al Laboratory (PNNL)), and Fabrizio Ferrandi (Polytechnic University of Mi
 lan)\n---------------------\nEvaluating Performance Portability of GPU Pro
 gramming Models\n\nMaintaining a single codebase that can achieve good per
 formance on a range of accelerator-based supercomputing platforms is of ex
 tremely high value for productive scientific application development. Howe
 ver, the large quantity of programming models available which claim to pro
 vide performance portab...\n\n\nJoshua H. Davis, Pranav Sivaraman, Isaac M
 inn, and Abhinav Bhatele (University of Maryland)\n---------------------\n
 High-Performance PMEM-Aware Collective I/Os\n\nCollective I/Os are widely 
 used to transform small non-contiguous accesses into large contiguous acce
 sses for parallel I/O optimization. The existing collective I/O techniques
  assume that computer memory is volatile. They are limited both by the siz
 e of the buffer, which must be small so data is not...\n\n\nKeegan Sanchez
  and Alex Gavin (Washington State University, Vancouver); Suren Byna (Ohio
  State University); Kesheng Wu (Lawrence Berkeley National Laboratory (LBN
 L)); and Xuechen Zhang (Washington State University, Vancouver)\n---------
 ------------\nTwo-Phase IO Enabling Large-Scale Performance Introspection\
 n\nNumerous sophisticated profiling and visualization tools have been deve
 loped to enable programmers to expose semantic information from their appl
 ication components.  However, effective and interactive exploration of the
  profiles of large-scale parallel programs remains a challenge due to the 
 high I/...\n\n\nKe Fan and Sidharth Kumar (University of Illinois, Chicago
 )\n---------------------\nPreserving Data Locality in Multidimensional Var
 iational Quantum Classification\n\nIn classical machine learning, the conv
 olution operation is leveraged in the eponymous class of convolutional neu
 ral networks (CNNs) capturing the spatial and/or temporal locality of mult
 idimensional input features. Preserving data locality allows CNN models to
  reduce the number of training paramete...\n\n\nMingyoung Jeng, Alvir Nobe
 l, Vinayak Jha, David Levy, Dylan Kneidel, Manu Chaudhary, Ishraq Islam, a
 nd Esam El-Araby (University of Kansas)\n\nTag: Artificial Intelligence/Ma
 chine Learning, Architecture and Networks, Heterogeneous Computing, I/O an
 d File Systems, Performance Measurement, Modeling, and Tools, Post-Moore C
 omputing, Programming Frameworks and System Software, Quantum Computing\n\
 nRegistration Category: Tech Program Reg Pass\n\nSession Chair: Dingwen Ta
 o (Institute of Computing Technology, Chinese Academy of Sciences; Univers
 ity of Chinese Academy of Sciences)
END:VEVENT
END:VCALENDAR