BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250626T234543Z
LOCATION:B302
DTSTART;TZID=America/New_York:20241118T093700
DTEND;TZID=America/New_York:20241118T100000
UID:submissions.supercomputing.org_SC24_sess756_ws_worksp101@linklings.com
SUMMARY:Performance Characterization and Provenance of Distributed Task-ba
 sed Workflows on HPC Platforms
DESCRIPTION:Amal Gueroudji (Argonne National Laboratory (ANL)); Chase Phel
 ps and Tanzima Z. Islam (Texas State University); Philip Carns, Shane Snyd
 er, Matthieu Dorier, and Robert Ross (Argonne National Laboratory (ANL)); 
 and Line Pouchard (Sandia National Laboratories)\n\nUnderstanding performa
 nce and provenance of task-based workflows poses significant challenges, p
 articularly in distributed configurations where resources are shared by mu
 ltiple applications. Task-based workflow management systems further compli
 cate performance predictability because of their dynamicity that subtly al
 ters task execution order from run to run.\nIn this paper we propose a lay
 ered characterization framework for performance and task provenance for Da
 sk.distributed workflows running on high-performance computing platforms. 
 It collects data from jobs, the workflow management system, and the operat
 ing system to aid in understanding the performance of these workflows. Our
  approach encompasses three main contributions: first, an extension of Das
 k.distributed to capture high-fidelity task provenance using Mochi data se
 rvices; second, the adaptation of the established HPC I/O characterization
  tool Darshan to gather high-fidelity I/O data, thereby enhancing the gran
 ularity of our analysis; and third, a framework to combine and process the
  collected data and provide helpful insights into performance characteriza
 tion and reproducibility.\n\nTag: Applications and Application Frameworks,
  Distributed Computing, Middleware and System Software\n\nRegistration Cat
 egory: Workshop Reg Pass\n\nSession Chairs: Silvina Caino-Lores (French In
 stitute for Research in Computer Science and Automation (INRIA)) and Anirb
 an Mandal\n\n
END:VEVENT
END:VCALENDAR