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:20211207T055357Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20211117T083000
DTEND;TZID=America/Chicago:20211117T170000
UID:submissions.supercomputing.org_SC21_sess279_rpost140@linklings.com
SUMMARY:Core-Idling on MPI Intra-Node Communication Channels for Energy Ef
 ficiency
DESCRIPTION:Posters, Research Posters\n\nCore-Idling on MPI Intra-Node Com
 munication Channels for Energy Efficiency\n\nKim, Jin, Byun\n\nBusy-waitin
 g used to implement parallel programming models can provide low latency bu
 t at the expense of energy. There have been several studies to enhance the
  energy efficiency of MPI libraries and applications by lowering the CPU s
 peed during busy-waiting or turning off CPU components. While most studies
  have focused on prediction-based schemes that determine to enter an energ
 y-saving mode, a redesign of the MPI library for better supports for suspe
 nding and resuming processes over multiple communication channels has been
  paid less attention. In this study, we suggest a signaling-based core-idl
 ing approach for energy efficient MPI over multiple communication channels
 . Our preliminary implementation supports intra-node communication channel
 s, such as shared memory and memory mapping. The performance measurement r
 esults show that our preliminary implementation can reduce the energy cons
 umption of NPB up to 12% in the undersubscribed case and 91% in the oversu
 bscribed case on a NUMA-based system.\n\nRegistration Category: Tech Progr
 am Reg Pass, Exhibit Hall Only
END:VEVENT
END:VCALENDAR