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:20260522T150124Z
LOCATION:D171
DTSTART;TZID=America/Chicago:20181113T140000
DTEND;TZID=America/Chicago:20181113T143000
UID:submissions.supercomputing.org_SC18_sess268_exforum129@linklings.com
SUMMARY:Enabling HPC and Deep Learning Workloads at Extreme Scale in the C
 loud
DESCRIPTION:Bill Bryce (Univa Corporation)\n\nIndependent research (Reuthe
 r et al., J. Parallel Distrib. Comput., 111, 2018, 76–92) underscores the 
 importance of efficient workload management: “For both supercomputers and 
 big data systems, the efficiency of the job scheduler represents a fundame
 ntal limit on the efficiency of the system.” However enabling efficiency a
 t extreme scale in the cloud, for workload management or other purposes, r
 equires sophisticated integration and automation that also scales. By deep
 ly integrating with AWS-specific APIs, the capabilities of this public-clo
 ud provider are fully leveraged via Navops Launch in a highly automated fa
 shion. As a compelling proof point, Navops Launch makes routine the scalin
 g of a compute cluster to more than 1,000,000 cores, across 55,000 heterog
 eneous spot instances spanning three availability zones. As a consequence,
  in demanding policy-based launching of cloud instances, heroics are no lo
 nger required to scale HPC and Deep Learning workloads to the extreme.\n\n
 Tag: Clouds and Distributed Computing, Deep Learning\n\nSession Chair: Ran
 dy Herban (Sylabs Inc)\n\n
END:VEVENT
END:VCALENDAR