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:20181221T160731Z
LOCATION:C141/143/149
DTSTART;TZID=America/Chicago:20181114T133000
DTEND;TZID=America/Chicago:20181114T140000
UID:submissions.supercomputing.org_SC18_sess177_pap229@linklings.com
SUMMARY:A Reference Architecture for Datacenter Scheduling: Design, Valida
 tion, and Experiments
DESCRIPTION:Paper\nClouds and Distributed Computing, Resource Management, 
 Scheduling, Tech Program Reg Pass\n\nA Reference Architecture for Datacent
 er Scheduling: Design, Validation, and Experiments\n\nAndreadis, Versluis,
  Mastenbroek, Iosup\n\nDatacenters act as cloud-infrastructure to stakehol
 ders across industry, government, and academia. To meet growing demand yet
  operate efficiently, datacenter operators employ increasingly more sophis
 ticated scheduling systems, mechanisms, and policies. Although many schedu
 ling techniques already exist, relatively little research has gone into th
 e abstraction of the scheduling process itself, hampering design, tuning, 
 and comparison of existing techniques. In this work, we propose a referenc
 e architecture for datacenter schedulers. The architecture follows five de
 sign principles: components with clearly distinct responsibilities, groupi
 ng of related components where possible, separation of mechanism from poli
 cy, scheduling as complex workflow, and hierarchical multi-scheduler struc
 ture. To demonstrate the validity of the reference architecture, we map to
  it state-of-the-art datacenter schedulers. We find scheduler-stages are c
 ommonly underspecified in peer-reviewed publications. Through trace-based 
 simulation and real-world experiments, we show underspecification of sched
 uler-stages can lead to significant variations in performance.
URL:https://sc18.supercomputing.org/presentation/?id=pap229&sess=sess177
END:VEVENT
END:VCALENDAR