BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230831T095754Z
LOCATION:Seehorn
DTSTART;TZID=Europe/Stockholm:20230626T163000
DTEND;TZID=Europe/Stockholm:20230626T183000
UID:submissions.pasc-conference.org_PASC23_sess161@linklings.com
SUMMARY:MS2E - Performance in I/O and Fault Tolerance for Scientific Appli
 cations
DESCRIPTION:Minisymposium\n\nThe interaction of modern scientific applicat
 ions in the exascale computing era with I/O resources is increasingly comp
 lex and important to application performance. Challenges in I/O performanc
 e are ubiquitous, from fault tolerance and resilience to coupled applicati
 on workflows to heterogeneous and task-based systems. The purpose of this 
 minisymposium is to facilitate a discussion of these challenges and recent
  novel research in high performance computing (HPC) addressing them. Topic
 s discussed will include research in the areas of fault tolerance, heterog
 eneous data, and workflow or coupled applications, with an emphasis on com
 puting resource, application, and data heterogeneity. We hope to enable a 
 conversation about how these techniques can be used across various applica
 tion use cases and thereby advance the state of the art in I/O performance
 . SNL is managed and operated by NTESS under DOE NNSA contract DE-NA000352
 5.\n\nTask-Level Resilience for Dynamically Generated Tasks under Work Ste
 aling in Clusters\n\nPermanent hardware failures of cluster nodes cause pr
 ocesses to abort and, if no precautions are taken, all previous compute re
 sults will be lost. Resilience can be achieved through checkpointing, whic
 h allows restarting applications from a saved state. However, writing chec
 kpoints to a file system ...\n\n\nClaudia Fohry (University of Kassel)\n--
 -------------------\nExploiting the Overlapping Challenges of Asynchronous
  Many Task Runtimes and Resilience for Integrated Control-Flow and Data Re
 siliency\n\nAs computing clusters grow in scale, so do the competing deman
 ds of resilience and performance. Contemporary high-performance runtimes a
 ccommodate for heterogeneous hardware, performance variability, and dynami
 c workload distributions; these challenges require gathering application c
 ontrol-flow and ...\n\n\nMatthew Whitlock and Nic Morales (Sandia National
  Laboratories) and Keita Teranishi (Oak Ridge National Laboratory)\n------
 ---------------\nScalable GPU-Accelerated Incremental Checkpointing of Spa
 rsely Updated Data\n\nCheckpointing large amounts of related data concurre
 ntly to stable storage is a common I/O pattern of many HPC applications in
  various scenarios including checkpoint-restart fault tolerance, coupled w
 orkflows that combine simulations with analytics, and adjoint computations
 . This pattern is challeng...\n\n\nMichela Taufer and Nigel Phillip Tan (U
 niversity of Tennessee), Bogdan Nicolae (Argonne National Laboratory), Jak
 ob Luettgau and Sanjukta Bhowmick (University of Tennessee), Keita Teranis
 hi (Oak Ridge National Laboratory), Nicolas Morales (Sandia National Labor
 atories), and Franck Cappello (Argonne National Laboratory)\n-------------
 --------\nOn the Role of Robust Staging Services for Extreme-Scale Workflo
 ws\n\nAs science applications target extreme scales (high-performance and 
 distributed computing) environments systems, dramatically increasing data 
 volumes and associated data management/IO costs are leading to in-situ app
 roaches to coupling and data processing and analysis. The resulting stagin
 g-based in...\n\n\nManish Parashar (Scientific Computing and Imaging Insti
 tute, University of Utah)\n\nDomain: Computer Science, Machine Learning, a
 nd Applied Mathematics 
\n\nSession Chair: Nicolas Morales (Sandia Nationa
 l Laboratories)
END:VEVENT
END:VCALENDAR