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:20211207T054759Z
LOCATION:240-241-242
DTSTART;TZID=America/Chicago:20211116T133000
DTEND;TZID=America/Chicago:20211116T140000
UID:submissions.supercomputing.org_SC21_sess150_pap408@linklings.com
SUMMARY:Bootstrapping In-Situ Workflow Auto-Tuning via Combining Performan
 ce Models of Component Applications
DESCRIPTION:Paper\n\nBootstrapping In-Situ Workflow Auto-Tuning via Combin
 ing Performance Models of Component Applications\n\nShu, Guo, Wozniak, Din
 g, Foster...\n\nIn an in-situ workflow, multiple components such as simula
 tion and analysis applications are coupled with streaming data transfers. 
 The multiplicity of possible configurations necessitates an auto-tuner for
  workflow optimization. Existing auto-tuning approaches are computationall
 y expensive because many configurations must be sampled by running the who
 le workflow repeatedly in order to train the auto-tuner surrogate model or
  otherwise explore the configuration space. To reduce these costs, we inst
 ead combine the performance models of component applications by exploiting
  the analytical workflow structure, selectively generating test configurat
 ions to measure and guide the training of a machine learning workflow surr
 ogate model. Because the training can focus on well-performing configurati
 ons, the resulting surrogate model can achieve high prediction accuracy fo
 r good configurations despite training with fewer total configurations. Ex
 periments with real applications demonstrate that our approach can identif
 y significantly better configurations than other approaches for a fixed co
 mputer time budget.\n\nTag: Reproducibility Badge, Algorithms, Application
 s, Performance\n\nRegistration Category: Tech Program Reg Pass\n\nReproduc
 ibility Badges: Artifact Available, Artifact Functional
END:VEVENT
END:VCALENDAR