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:20181221T160730Z
LOCATION:D173
DTSTART;TZID=America/Chicago:20181113T160000
DTEND;TZID=America/Chicago:20181113T163000
UID:submissions.supercomputing.org_SC18_sess274_exforum108@linklings.com
SUMMARY:Analytic Based Monitoring of High Performance Computing Applicatio
 ns
DESCRIPTION:Exhibitor Forum\nData Analytics, System Software\n\nAnalytic B
 ased Monitoring of High Performance Computing Applications\n\nLeinberger\n
 \nThe complexity of High Performance Computing (HPC) systems and the innat
 e premium on system efficiency necessitate the use of automated tools to m
 onitor not only system-level health and status, but also job performance. 
 Current vendor-provided and third party monitoring tools, such as Nagios o
 r Ganglia, enable system-level monitoring using features that reflect the 
 state of system resources. None of those tools, however, are designed to d
 etermine the health and status of a user’s application, or job, as it exec
 utes.  \n\nThis presentation introduces the concept of job-level, analytic
 s-based monitoring using system features external to the job, like those r
 eported by Ganglia. Preliminary results show these features contain suffic
 ient information content to characterize key behaviors of an executing job
  when incorporated into a job-specific, application-neutral analytic model
 ; transitions between computational phases, onset of a load imbalance, and
  anomalous activity on a compute node may each be detected using this appr
 oach.
URL:https://sc18.supercomputing.org/presentation/?id=exforum108&sess=sess2
 74
END:VEVENT
END:VCALENDAR