BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260202T201803Z
LOCATION:230
DTSTART;TZID=America/Chicago:20251117T163500
DTEND;TZID=America/Chicago:20251117T164000
UID:submissions.supercomputing.org_SC25_sess202_ws_pdswwip102@linklings.co
 m
SUMMARY:Accelerating Exascale Scientific Discovery via In-Situ and In-Tran
 sit Data Analytics in HPC
DESCRIPTION:Vijayalakshmi Saravanan (University of Texas at Tyler, Univers
 ity of Texas at Dallas); Sai Karthik Navuluru (UTD); and Khaled Z. Ibrahim
  (Lawrence Berkeley National Laboratory (LBNL))\n\nThe rapid growth of mul
 timodal data from large-scale simulations and experimental instruments is 
 overwhelming traditional storage and analysis workflows. Post hoc, disk-ba
 sed methods suffer from latency, bandwidth bottlenecks, and inefficient re
 source use, slowing scientific insight. This work explores a hybrid in-sit
 u and in-transit framework that embeds computation within the memory and s
 torage hierarchy of HPC systems. In-situ processing performs filtering, re
 duction, or analysis directly at the data source using node-local memory a
 nd accelerators. In-transit processing complements this by leveraging inte
 rmediate layers such as burst buffers or dedicated resources for asynchron
 ous analytics, balancing simulation and analysis.\nOur architecture integr
 ates Apache Ignite’s in-memory data grid with Apache Spark’s distributed c
 omputing and containerized microservices to enable real-time ingestion, fu
 sion, and ML-driven analysis. Our preliminary results show reduced latency
 , efficient CPU–memory utilization, and strong scalability. Case studies o
 n NWChem molecular dynamics and E3SM climate simulations demonstrate adapt
 ability across domains, advancing data-aware, exascale-class discovery.\n\
 nTag: Data Analytics, High Performance I/O, Storage, Archive, & File Syste
 ms, Storage\n\nRecording: Livestreamed, Recorded\n\nRegistration Category:
  Technical Program Reg Pass, Workshop Reg Pass\n\nSession Chairs: Suren By
 na (The Ohio State University, Lawrence Berkeley National Laboratory (LBNL
 )); Anthony Kougkas (Illinois Institute of Technology, Argonne National La
 boratory (ANL)); Sarah Neuwirth (Johannes Gutenberg University Mainz); Ric
 ardo Macedo (INESC TEC, University of Minho); Jay Lofstead (Sandia Nationa
 l Laboratories, University of New Mexico); and Dean Hildebrand (Google LLC
 )\n\n
END:VEVENT
END:VCALENDAR