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:20200129T163557Z
LOCATION:708
DTSTART;TZID=America/Denver:20191117T154500
DTEND;TZID=America/Denver:20191117T160000
UID:submissions.supercomputing.org_SC19_sess112_pec250@linklings.com
SUMMARY:Empowering Agroecosystem Modeling with HTC Scientific Workflows: T
 he Cycles Model Use Case
DESCRIPTION:Workshop\n\nEmpowering Agroecosystem Modeling with HTC Scienti
 fic Workflows: The Cycles Model Use Case\n\nFerreira da Silva, Mayani, Shi
 , Kemanian, Rynge...\n\nScientific workflows have enabled large-scale scie
 ntific computations and data analysis, and lowered the entry barrier for p
 erforming computations in distributed heterogeneous platforms (e.g., HTC a
 nd HPC). In spite of impressive achievements to date, large-scale modeling
 , simulation, and data analytics in the long-tail still face several chall
 enges such as efficient scheduling and execution of large-scale workflows 
 (O(10^6)) with very short-running tasks (few seconds). While the current t
 rend to support next-generation workflows on leadership class machines hav
 e gained much attention in the past years, at the other end of the spectru
 m scientific workflows from the long-tail science have become larger and r
 equire processing massive volumes of data. In this paper, we report on our
  experience in designing and implementing an HTC workflow for agroecosyste
 m modeling. We leverage well-known (task clustering and co-scheduling) and
  emerging (hierarchical workflows and containers) workflow optimization te
 chniques to make the workflow planning problem tractable, and maximize res
 ource utilization and the degree of task parallelism. Experimental results
 , via the implementation of a use case, show that by strategically combini
 ng the above strategies and defining an appropriate set of optimization pa
 rameters, the overall workflow makespan can be improved by 3.5 orders of m
 agnitude when compared to a regular (non-optimized) execution of the workf
 low.\n\nTag: Workshop Reg Pass, Extreme Scale Computing, Scalable Computin
 g, Scientific Workflows\n\nRegistration Category: Workshop Reg Pass, Extre
 me Scale Computing, Scalable Computing, Scientific Workflows
URL:https://sc19.supercomputing.org/presentation/?id=pec250&sess=sess112
END:VEVENT
END:VCALENDAR