BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160557Z
LOCATION:Track 9
DTSTART;TZID=America/New_York:20201111T143000
DTEND;TZID=America/New_York:20201111T150000
UID:submissions.supercomputing.org_SC20_sess196_ws_works106@linklings.com
SUMMARY:Adaptive Optimizations for Stream-Based Workflows
DESCRIPTION:Workshop\n\nAdaptive Optimizations for Stream-Based Workflows\
 n\nLiang, Filgueira, Yan\n\nThis work presents three new adaptive optimiza
 tion techniques to maximize the performance of dispel4py workflows.  dispe
 l4py is a parallel Python-based stream-oriented dataflow framework that ac
 ts as a bridge to existing parallel programming frameworks like MPI or Pyt
 hon multiprocessing.  When a user runs a dispel4py workflow, the original 
 framework performs a fixed workload distribution among the processes avail
 able for the run.  This allocation does not take into account workflows’ f
 eatures, which can cause scalability issues, especially for data-intensive
  scientific workflows.  Therefore, our aim is to improve the performance o
 f dispel4py workflows by testing different workload strategies that automa
 tically adapt to workflows.  For achieving this objective, we have impleme
 nted three new techniques, called Naive Assignment, Staging, and Dynamic S
 cheduling.  The evaluations show that our proposed techniques have signifi
 cantly improved the performance of the original dispel4py framework.\n\nTa
 g: Scientific Computing, Workflows\n\nRegistration Category: Workshop Reg 
 Pass
END:VEVENT
END:VCALENDAR