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:20211207T055404Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211114T090100
DTEND;TZID=America/Chicago:20211114T100000
UID:submissions.supercomputing.org_SC21_sess430_misc303@linklings.com
SUMMARY:ExaMPI Invited Talk: Designing High-Performance and Scalable Middl
 eware for HPC and AI: Challenges and Opportunities
DESCRIPTION:Workshop\n\nExaMPI Invited Talk: Designing High-Performance an
 d Scalable Middleware for HPC and AI: Challenges and Opportunities\n\nPand
 a\n\nThis talk will focus on challenges and opportunities in designing mid
 dleware for HPC, AI (Deep/Machine Learning), and Data Science for On-premi
 se HPC and Cloud systems with advances in networking and accelerator techn
 ologies. For the HPC domain, we will discuss about the challenges in desig
 ning runtime environments for MPI+X programming models by taking into acco
 unt support for multi-core systems (Xeon, ARM and OpenPower), high-perform
 ance networks (InfiniBand and RoCE), GPUs (including GPUDirect RDMA), and 
 emerging BlueField-2 DPUs. Features, sample performance numbers and best p
 ractices of using MVAPICH2 libraries (http://mvapich.cse.ohio-state.edu)wi
 ll be presented. For the Deep/Machine Learning domain, we will focus on MP
 I-driven solutions (http://hidl.cse.ohio-state.edu) to extract performance
  and scalability for popular Deep Learning frameworks (TensorFlow and PyTo
 rch) and large out-of-core models. Accelerating Deep Learning applications
  with Bluefield-2 DPUs will also be presented. MPI-driven solutions to acc
 elerate data science applications like Dask (http://hibd.cse.ohio-state.ed
 u) will be highlighted. Finally, we will outline the challenges and experi
 ences in deploying this middleware to the HPC cloud environments for Azure
 , AWS, and Oracle.\n\nTag: Online Only, Extreme Scale Computing, Parallel 
 Programming Languages and Models, Performance\n\nRegistration Category: Wo
 rkshop Reg Pass
END:VEVENT
END:VCALENDAR