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:20211207T055402Z
LOCATION:Online
DTSTART;TZID=America/Chicago:20211115T112000
DTEND;TZID=America/Chicago:20211115T114500
UID:submissions.supercomputing.org_SC21_sess342_ws_worksp105@linklings.com
SUMMARY:Not All Tasks Are Created Equal: Adaptive Resource Allocation for 
 Heterogeneous Tasks in Dynamic Workflows
DESCRIPTION:Workshop\n\nNot All Tasks Are Created Equal: Adaptive Resource
  Allocation for Heterogeneous Tasks in Dynamic Workflows\n\nPhung, Thain, 
 Chard, Ward\n\nUsers running dynamic workflows in distributed systems usua
 lly have inadequate expertise to correctly size the allocation of resource
 s (cores, memory, disk) to each task due to the difficulty in uncovering t
 he obscure yet important correlation between tasks and their resource cons
 umption. Thus, users typically pay little attention to this problem of all
 ocation sizing and either simply apply an error-prone upper bound of resou
 rce allocation to all tasks, or delegate this responsibility to underlying
  distributed systems, resulting in substantial waste from allocated yet un
 used resources. In this paper, we will first show that tasks performing di
 fferent work may have significantly different resource consumption. We wil
 l then show that exploiting the heterogeneity of tasks is a desirable way 
 to reveal and predict the relationship between tasks and their resource co
 nsumption, reduce waste from resource mis-allocation, increase tasks’ cons
 umption efficiency, and incentivize users’ cooperation. We have developed 
 two info-aware allocation strategies capitalizing on this characteristic a
 nd will show their effectiveness through simulations on two modern applica
 tions with dynamic workflows and five synthetic datasets of resource consu
 mption. Our results show that info-aware strategies can cut down up to 98.
 7% of the total waste incurred by a best-effort strategy, and increase the
  efficiency in resource consumption of each task on average anywhere up to
  93.9%.\n\nTag: Online Only, Cloud and Distributed Computing, Scientific C
 omputing, Workflows\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR