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:20181221T160728Z
LOCATION:D168
DTSTART;TZID=America/Chicago:20181112T121000
DTEND;TZID=America/Chicago:20181112T121500
UID:submissions.supercomputing.org_SC18_sess140_ws_isav113@linklings.com
SUMMARY:Leveraging Scalable Event Distribution to Enable Data-Driven In Si
 tu Scientific Workflows
DESCRIPTION:Workshop\nData Analytics, Data Management, Visualization, Work
 shop Reg Pass\n\nLeveraging Scalable Event Distribution to Enable Data-Dri
 ven In Situ Scientific Workflows\n\nWang, Simonet, Subedi, Davis, Parashar
 \n\nNovel event-driven workflow systems have been effectively used to incr
 ease the performance of large-scale scientific applications by removing mo
 st of the implicit synchronization required to orchestrate distributed tas
 ks. However, these event-driven workflow systems, by focusing only on even
 ts related to the completion of tasks and data transfers, fail to address 
 the dynamic and irregular workflows that require fine adaptation of the ex
 ecution to the environment, faults, and to partial results from the applic
 ation itself.\n\nIn this article, we explore the idea of a programming mod
 el for irregular and dynamic workflows that is not only based on task-rela
 ted events, but also on the intermediate data produced the tasks. We conte
 nd that compared to traditional workflow execution systems this technique 
 will ease development, increase flexibility and performance by removing im
 plicit synchronization and automating previously tedious tasks related to 
 workflow steering. We identify the classes of workflows that will benefit 
 the most from this model and discuss design considerations for future impl
 ementations. In particular, we discuss how novel in-situ analysis techniqu
 es can be leveraged to implement a workflow system based on events of vari
 ous natures and origins, from the infrastructure to the intermediate data 
 while a workflow is running.
URL:https://sc18.supercomputing.org/presentation/?id=ws_isav113&sess=sess1
 40
END:VEVENT
END:VCALENDAR