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:20260422T000611Z
LOCATION:505
DTSTART;TZID=America/Denver:20231112T140000
DTEND;TZID=America/Denver:20231112T173000
UID:submissions.supercomputing.org_SC23_sess428@linklings.com
SUMMARY:6th International Workshop on Emerging Parallel Distributed Runtim
 e Systems and Middleware
DESCRIPTION:DeepSpeed4Science: Enabling Future Large-Scale Scientific Disc
 overy through Sophisticated AI System Technologies\n\nShuaiwen Leon Song (
 Microsoft Corporation)\n---------------------\nDask-Extended External Task
 s for HPC/ML In Transit Workflows\n\nIn situ workflows are inescapable to 
 fully leverage exascale architectures. They can however be complex to buil
 d as simulation and data analytics come from two different software ecosys
 tems with their own paradigms. We extend deisa by introducing the concept 
 of external tasks to support the descript...\n\n\nAmal Gueroudji (Argonne 
 National Laboratory (ANL)), Julien Bigot (Atomic Energy and Alternative En
 ergies Commission (CEA)), Bruno Raffin (INRIA), and Robert Ross (Argonne N
 ational Laboratory (ANL))\n---------------------\nIPDRM’2023 – Welcome\n\n
 Barbara Chapman (Hewlett Packard Enterprise (HPE))\n---------------------\
 nA gem5 Implementation of the Sequential Codelet Model: Reducing Overhead 
 and Expanding the Software Memory Interface\n\nModern tasking models defin
 e applications in a fine-grained manner that necessitates lower overhead p
 er segment of computation. While previous work has seen implementations of
  hardware support for tasking models, many lack the support required by he
 terogeneity and fall short of expanding memory inte...\n\n\nDawson Fox (Un
 iversity of Delaware, Argonne National Laboratory (ANL)); Jose Monsalve Di
 az (Argonne National Laboratory (ANL)); and Xiaoming Li (University of Del
 aware)\n---------------------\nIPDRM’2023 – Afternoon Break\n-------------
 --------\nMPI-xCCL: A Portable MPI Library over Collective Communication L
 ibraries for Various Accelerators\n\nThe evolution of high-performance com
 puting toward diverse accelerators, including NVIDIA, AMD, Intel GPUs, and
  Habana Gaudi Accelerators, demands a user-friendly and efficient utilizat
 ion of these technologies. While both GPU-aware MPI libraries and vendor-s
 pecific communication libraries cater to ...\n\n\nChen-Chun Chen, Kawthar 
 Shafie Khorassani, Pouya Kousha, Qinghua Zhou, Jinghan Yao, Hari Subramoni
 , and Dhabaleswar K. Panda (Ohio State University)\n---------------------\
 nEnabling Large Dynamic Neural Network Training with Learning-Based Runtim
 e Memory Management\n\nDynamic neural network (DyNN) enables high computat
 ional efficiency and strong representation capability. However, training D
 yNN can face a memory capacity problem because of increasing model size or
  limited GPU memory capacity. Managing tensors to save GPU memory is chall
 enging, because of the dyna...\n\n\nDong Li (University of California, Mer
 ced)\n---------------------\nClosing Remarks\n\nOceane Bel (Pacific Northw
 est National Laboratory (PNNL))\n---------------------\nHPC Software Scali
 ng for ML Using CXL 3.0 GFAM\n\nTraditional HPC systems rely on balanced s
 oft scaling, which adjusts the compute-to-memory ratio according to the wo
 rkload. However, this approach is challenged by Machine Learning applicati
 ons, especially Large Language Model (LLM) workloads, which demand much mo
 re memory than compute. This leads t...\n\n\nPatrick Estep (Micron Technol
 ogy Inc)\n\nTag: Distributed Computing, Middleware and System Software, Ru
 ntime Systems\n\nRegistration Category: Workshop Reg Pass\n\nSession Chair
 s: Barbara Chapman (Hewlett Packard Enterprise (HPE), Stony Brook Universi
 ty); Joseph Manzano (Pacific Northwest National Laboratory (PNNL)); Shirle
 y Moore (University of Texas at El Paso); EunJung (EJ) Park (Qualcomm); an
 d Joshua Suetterlein (Pacific Northwest National Laboratory (PNNL))
END:VEVENT
END:VCALENDAR