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:20260422T000612Z
LOCATION:605
DTSTART;TZID=America/Denver:20231113T140000
DTEND;TZID=America/Denver:20231113T173000
UID:submissions.supercomputing.org_SC23_sess453@linklings.com
SUMMARY:ExaMPI: Workshop on Exascale MPI
DESCRIPTION:Optimizing Irregular Communication with Neighborhood Collectiv
 es and Locality-Aware Parallelism\n\nIrregular communication limits both p
 erformance and scalability of parallel applications. Typically, it is impl
 emented as point-to-point, and optimizations are integrated into the appli
 cation, lacking portability. Optimization of point-to-point messages withi
 n MPI is difficult, as the interface only...\n\n\nGerald Collom (Universit
 y of New Mexico), Rui Peng Li (Lawrence Livermore National Laboratory (LLN
 L)), and Amanda Bienz (University of New Mexico)\n---------------------\nE
 xaMPI: Workshop on Exascale MPI\n\nThe aim of this workshop is to bring to
 gether researchers and developers to present and discuss innovative algori
 thms and concepts in the message passing programming model and to create a
  forum for open and potentially controversial discussions on the future of
  MPI in the exascale era and beyond.\n\n\nRyan Grant (Queen's University, 
 Canada); Matthew Dosanjh (Sandia National Laboratories); Purushotham Banga
 lore (University of Alabama); Anthony Skjellum (University of Tennessee, C
 hattanooga); Amanda Bienz (University of New Mexico); and William Schonbei
 n (Sandia National Laboratories)\n---------------------\nUsing Mixed-Radix
  Decomposition to Enumerate Computational Resources of Deeply Hierarchical
  Architectures\n\nCurrent HPC architectures are deeply hierarchical (racks
 , nodes, sockets, NUMA domains, caches, ...), and the mapping of MPI proce
 sses to cores can significantly influence application performance. To stud
 y hierarchy effects on MPI application performance, we propose a procedure
  for expressing mappin...\n\n\nPhilippe Swartvagher, Sascha Hunold, Jesper
  Larsson Träff, and Ioannis Vardas (Technical University Wien (Vienna Univ
 ersity of Technology))\n---------------------\nDistinguished Speaker:  GPU
  Centric Communication – Is MPI Missing Out?\n\nJames Dinan (NVIDIA Corpor
 ation)\n---------------------\nExaMPI – Afternoon Break\n-----------------
 ----\nWelcome and Introduction\n\nRyan Grant\n---------------------\nA Sta
 tistical Analysis of HPC Network Tuning\n\nDistributed scientific applicat
 ions run on a complex stack of soft- ware and network technologies. Each l
 ayer has configuration options for tuning performance. These can range fro
 m protocol thresh- olds to algorithmic changes for collectives. Micro-benc
 hmarks are a common methodology to evaluate the...\n\n\nDonald Kruse, Whit
  Schonbein, and Matthew G. F. Dosanjh (Sandia National Laboratories)\n----
 -----------------\nEmbedding Rust within Open MPI\n\nThe Message-Passing I
 nterface (MPI) requires implementations that are able to adapt to new hard
 ware and architectures while ensuring correctness and usability. The most 
 widely used MPI implementations, however, are written in older programming
  languages that can lead to memory-unsafe code with poor ...\n\n\nJake Tro
 nge and Howard Pritchard (Los Alamos National Laboratory (LANL))\n--------
 -------------\nOpenSHMEM Queues:  An Abstraction for Enhancing Message Rat
 e, Bandwidth Utilization, and Reducing Tail Latency in OpenSHMEM Applicati
 ons\n\nVishwanath Venkatesan and Manjunath Gorentla Venkata (NVIDIA Corpor
 ation)\n---------------------\nEfficient Data Redistribution for Malleable
  Applications\n\nProcess malleability can be defined as the ability of a d
 istributed MPI parallel job to change the number of processes on--the--fly
  without stopping its execution, reallocating the compute resources origin
 ally assigned to the job, and without storing application data to disk.  M
 PI malleability consi...\n\n\nIker Martín Álvarez, José I. Aliaga, and Mar
 ibel Castillo (Universitat Jaume I) and Sergio Iserte (Barcelona Supercomp
 uting Center (BSC))\n\nTag: Exascale, Message Passing, Programming Framewo
 rks and System Software\n\nRegistration Category: Workshop Reg Pass\n\nSes
 sion Chairs: Purushotham Bangalore (The University of Alabama); Amanda J. 
 Bienz (University of New Mexico); Matthew G. F. Dosanjh (Sandia National L
 aboratories); Ryan Grant (Queen's University, Canada; Power API); William 
 Schonbein (Sandia National Laboratories); and Anthony Skjellum (Tennessee 
 Technological University, ASCEND-TNTECH)
END:VEVENT
END:VCALENDAR