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:20181221T160904Z
LOCATION:C2/3/4 Ballroom
DTSTART;TZID=America/Chicago:20181113T083000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess325_spost132@linklings.com
SUMMARY:Modeling Single-Source Shortest Path Algorithm Dynamics to Control
  Performance and Power Tradeoffs
DESCRIPTION:ACM Student Research Competition, Poster\nTech Program Reg Pas
 s, Exhibits Reg Pass\n\nModeling Single-Source Shortest Path Algorithm Dyn
 amics to Control Performance and Power Tradeoffs\n\nKaramati, Young, Vuduc
 \n\nThis work presents a new methodology to improve the performance of par
 allel algorithms by tuning the amount of available parallelism for executi
 on throughout the runtime. As such, we expose key parameters controlling t
 he performance and parallelism of the algorithm and build a software-based
  controller with the objective of maintaining the optimal performance. Our
  controller allows for tuning the level of parallelism executed in each ti
 me epoch to optimize for performance while preserving power usage. More sp
 ecifically, our experimental evaluation focuses on a tunable variation of 
 a GPU-based delta-stepping algorithm for computing the single-source short
 est path (SSSP); As the available parallelism for the delta-stepping SSSP 
 is highly irregular and strongly input-dependent, our extensive experiment
 s show that average power can be reduced while average parallelism is incr
 eased. This increase in average parallelism provides substantial energy sa
 vings, independent of the hardware.
URL:https://sc18.supercomputing.org/presentation/?id=spost132&sess=sess325
END:VEVENT
END:VCALENDAR

