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:20181221T160730Z
LOCATION:C140/142
DTSTART;TZID=America/Chicago:20181113T163000
DTEND;TZID=America/Chicago:20181113T170000
UID:submissions.supercomputing.org_SC18_sess182_pap311@linklings.com
SUMMARY:Mitigating Inter-Job Interference Using Adaptive Flow-Aware Routin
 g
DESCRIPTION:Paper\nNetworks, Resource Management, Scheduling, State of the
  Practice, System Software, Tech Program Reg Pass, BSP Finalist\n\nMitigat
 ing Inter-Job Interference Using Adaptive Flow-Aware Routing\n\nSmith, Cro
 mey, Lowenthal, Domke, Jain...\n\nOn most high performance computing platf
 orms, applications share network resources with other jobs running concurr
 ently on the system.  Inter-job network interference can have a significan
 t impact on the performance of communication-intensive applications, and n
 o satisfactory solutions yet exist for mitigating this degradation.\n\nIn 
 this paper, we analyze network congestion caused by multi-job workloads on
  two production systems that use popular network topologies---fat-tree and
  dragonfly. For each system, we establish a regression model to relate net
 work hotspots to application performance degradation, showing that current
  routing strategies are insufficient to load-balance network traffic and m
 itigate interference on production systems.  We then propose an alternativ
 e type of adaptive routing strategy, which we call adaptive flow-aware rou
 ting.  We implement a prototype of our strategy, and tests on the fat-tree
  system show up to a 46% improvement in job run time when compared to the 
 default routing.
URL:https://sc18.supercomputing.org/presentation/?id=pap311&sess=sess182
END:VEVENT
END:VCALENDAR

