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:20260422T000604Z
LOCATION:DEF Concourse
DTSTART;TZID=America/Denver:20231115T100000
DTEND;TZID=America/Denver:20231115T170000
UID:submissions.supercomputing.org_SC23_sess303_rpost184@linklings.com
SUMMARY:Graph Based Anomaly Detection in Chimbuko:  Feasible or Fallible?
DESCRIPTION:Chase Phelps, Ankur Lahiry, and Tanzima Z. Islam (Texas State 
 University) and Christopher Kelly (Brookhaven National Laboratory)\n\nPerf
 ormance anomaly detection can aid in discovering algorithmic inefficiencie
 s or hardware issues in an application’s environment. The Chimbuko framewo
 rk monitors large-scale workflow applications in real-time and identifies 
 function executions which deviate from accumulated statistics (performance
  anomalies). Performance anomalies across runs correlate with variation in
  execution times of an application; quicker resolution of performance anom
 alies caused by hardware issues improves cluster performance. Anomalous an
 d normal executions are stored as events in Chimbuko. In this study, we in
 vestigate the applicability of graph-based deep learning methods for anoma
 ly classification. We hypothesize that transforming data into a graph will
  allow correlations to be modeled, thus allowing graph-based methods to le
 arn embeddings that can improve the effectiveness of downstream anomaly cl
 assification tasks. Our evaluations demonstrate that the graph-based metho
 ds yield up to 95% accuracy and outperform a state-of-the-art gradient-bas
 ed method. Moreover, we provide an explanation of the classification model
 ’s decision-making process through explainable AI techniques.\n\nTag: Arti
 ficial Intelligence/Machine Learning, Architecture and Networks, Heterogen
 eous Computing, I/O and File Systems, Performance Measurement, Modeling, a
 nd Tools, Post-Moore Computing, Programming Frameworks and System Software
 , Quantum Computing\n\nRegistration Category: Tech Program Reg Pass, Exhib
 its Reg Pass\n\n
END:VEVENT
END:VCALENDAR