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:20260422T000618Z
LOCATION:710
DTSTART;TZID=America/Denver:20231112T140000
DTEND;TZID=America/Denver:20231112T173000
UID:submissions.supercomputing.org_SC23_sess427@linklings.com
SUMMARY:Fourth International Symposium on Checkpointing for Supercomputing
  (SuperCheck-SC23)
DESCRIPTION:Implementation-Oblivious Transparent Checkpoint-Restart for MP
 I\n\nThis work presents experience with traditional use cases of checkpoin
 ting on a novel platform.  A single codebase (MANA) transparently checkpoi
 nts production workloads for major, available MPI implementations: "develo
 p once, run everywhere''.  The new platform allows application developers 
 to compil...\n\n\nYao Xu, Leonid Belyaev, and Twinkle Jain (Northeastern U
 niversity); Derek Schafer (University of New Mexico); Anthony Skjellum (Te
 nnessee Tech University); and Gene Cooperman (Northeastern University)\n--
 -------------------\nLightning Talk:  Datastates for Debugging – Using Pro
 ductive Checkpointing for Improved Debugging\n\nDebuggers are powerful and
  productive tools to understand and fix correctness issues in parallel pro
 grams.  Tools like `rr` and `gdb` even introduce capabilities for “reverse
  time” debugging allowing the user to step backward through the states of 
 the program.  However, these tools were ...\n\n\nRobert Underwood (Argonne
  National Laboratory (ANL))\n---------------------\nLightning Talk:  Diasp
 ora – Resilient Event Processing for Irregular, Distributed Scientific App
 lications\n\nModern science increasingly requires the coordinated use of a
 dvanced computing, networking, instruments, and experimental facilities: c
 ollectively, research infrastructure. This infrastructure reaches from HPC
  systems to high-data-rate instruments and less well-connected edge system
 s, and also encom...\n\n\nJustin Wozniak (Argonne National Laboratory (ANL
 ))\n---------------------\nAI-Augmented SWARM Based Resilience for Integra
 te Research Infrastructures\n\nThe community spent a dozen years developin
 g production-level checkpointing techniques, such as VeloC, capable of cap
 turing and saving extreme volumes of data with negligible overhead for the
  exascale scientific parallel application executions. A novel category of 
 systems will emerge within the next...\n\n\nFranck Cappello (Argonne Natio
 nal Laboratory (ANL))\n---------------------\nWelcome to SuperCheck-SC23\n
 \nRebecca Hartman-Baker (NERSC at LBNL (Lawrence Berkeley National Laborat
 ory))\n---------------------\nSuperCheck-SC23 – Afternoon Break\n---------
 ------------\nAsynchronous Multi-Level Checkpointing: An Enabler of Reprod
 ucibility using Checkpoint History Analytics\n\nHigh-performance computing
  applications are increasingly integrating checkpointing libraries for rep
 roducibility analytics. However, capturing an entire checkpoint history fo
 r reproducibility study faces the challenges of high-frequency checkpointi
 ng across thousands of processes. As a result, the r...\n\n\nKevin Assogba
  (Rochester Institute of Technology), Bogdan Nicolae (Argonne National Lab
 oratory (ANL)), Huub Van Dam (Brookhaven National Laboratory), and M. Must
 afa Rafique (Rochester Institute of Technology)\n---------------------\nLi
 ghtning Talk:  Inherent Checkpointing Properties of Nested Parallelism\n\n
 Our discovery was a complete accident. We worked on nested parallelism in 
 OpenMP for easier work decomposition and better work balancing:\n\n#pragma
  omp parallel num_threads(N)\n{\n    #pragma omp parallel num_threads(M)\n
     {\n        #pragma omp parallel num_threads(K)\n        {\n        }\n
     }\n}\n\nCrea...\n\n\nStanislav Bratanov (Intel Corporation)\n---------
 ------------\nCheckpoint/Restart for CUDA Kernels\n\nIn HPC clusters, it h
 as become common to employ Checkpoint/Restart, that is, saving the executi
 on state of applications in order to restore their computational progress 
 at a later point in time. The benefits of this technique for clusters incl
 ude more flexibility when reacting to changing workloads ...\n\n\nNiklas E
 iling, Stefan Lankes, and Antonello Monti (RWTH Aachen University)\n------
 ---------------\nLightning Talk:  Toward Efficient Asynchronous Checkpoint
 ing for Large-Language Models\n\nLarge-language models (LLMs) have been ra
 pidly and widely adopted across research, academia, and enterprises for ex
 ploring different endeavors ranging from scientific and educational pursui
 ts to financial and legal assistance. Unsurprisingly, training such sophis
 ticated LLMs requires large-scale inf...\n\n\nAvinash Maurya (Rochester In
 stitute of Technology)\n---------------------\nLightning Talk:  Trade-Offs
  For Developing File Aggregated I/O For Asynchronous Checkpointing\n\nAsyn
 chronous checkpoint-restart (C/R) has become popular in recent years for i
 ts ability to checkpoint alongside the application. One implementation  is
  VELOC, which quickly checkpoints to a local storage device, and then flus
 hes the checkpoints to the PFS concurrently with the application. VELOC na
 ...\n\n\nMikaila Gossman (Clemson University)\n---------------------\nLigh
 tning Talk:  Update on Checkpointing and Localized Recovery for Nested For
 k-Join Programs\n\nAn important use case for checkpointing is the protecti
 on of programs against permanent node failures in clusters. Ideally, check
 points should be written transparently, i.e., without the need to change t
 he program; and efficiently, i.e., they should include only relevant data.
  Other frequently desir...\n\n\nClaudia Fohry (University of Kassel, Germa
 ny)\n\nTag: Fault Handling and Tolerance\n\nRegistration Category: Worksho
 p Reg Pass\n\nSession Chairs: Gene Cooperman (Northeastern University); Do
 nglai Dai (Advanced Micro Devices (AMD)); Rebecca Hartman-Baker (National 
 Energy Research Scientific Computing Center (NERSC), Lawrence Berkeley Nat
 ional Laboratory (LBNL)); and Bogdan Nicolae (Argonne National Laboratory 
 (ANL), Illinois Institute of Technology)
END:VEVENT
END:VCALENDAR