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:20181221T160727Z
LOCATION:D170
DTSTART;TZID=America/Chicago:20181112T090500
DTEND;TZID=America/Chicago:20181112T092500
UID:submissions.supercomputing.org_SC18_sess169_ws_daac106@linklings.com
SUMMARY:Contention-Aware Container Placement Strategy for Docker Swarm
DESCRIPTION:Workshop\nHPC Center Planning and Operations, Heterogeneous Sy
 stems, Scheduling, Scientific Computing, State of the Practice, Workshop R
 eg Pass, Containers, Datacenter\n\nContention-Aware Container Placement St
 rategy for Docker Swarm\n\nChiang\n\nContainerization technology utilizes 
 operating system level virtualization to package applications so they can 
 run with required libraries and are isolated from other processes on the s
 ame host.  Lightweight and quick deployment make containers popular in man
 y data centers. Running distributed applications in data centers usually i
 nvolves multiple clusters of machines. Docker Swarm is a container orchest
 ration tool for managing a cluster of Docker containers and their hosts. H
 owever, Docker Swarm's scheduler does not consider resource utilization wh
 en placing containers in a cluster. This paper first investigated performa
 nce interference in container clusters. Our experimental study showed that
  distributed applications' performance can be degraded by about 15% when c
 o-located with other containers which aggressively consume resources. We t
 hen proposed a new scheduler to improve performance while keeping high res
 ource utilization. The experimental results demonstrated that the proposed
  prototype can effectively improve distributed applications' performance b
 y up to 3.13%.
URL:https://sc18.supercomputing.org/presentation/?id=ws_daac106&sess=sess1
 69
END:VEVENT
END:VCALENDAR