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:20211207T055405Z
LOCATION:231-232
DTSTART;TZID=America/Chicago:20211114T110000
DTEND;TZID=America/Chicago:20211114T113000
UID:submissions.supercomputing.org_SC21_sess427_ws_canopie107@linklings.co
 m
SUMMARY:It’s a Scheduling Affair: GROMACS in the Cloud with the KubeFlux S
 cheduler
DESCRIPTION:Workshop\n\nIt’s a Scheduling Affair: GROMACS in the Cloud wit
 h the KubeFlux Scheduler\n\nMisale, Drocco, Milroy, Arango Gutierrez, Herb
 ein...\n\nIn this work, we address the problem of running HPC workloads ef
 ficiently on Kubernetes clusters. To do so, we compare the Kubernetes' def
 ault scheduler with KubeFlux, a Kubernetes plug-in scheduler built on the 
 Flux graph-based scheduler, on a 34-node Red Hat OpenShift cluster on IBM 
 Cloud. We detail how scheduling can affect the performance of GROMACS, a w
 ell-known HPC application, and we show that KubeFlux can improve its perfo
 rmance through better pod scheduling. In our tests, KubeFlux demonstrates 
 the tendency to limit the number of subnets spanned by a job and the maxim
 um number of pods per node, translating to a >2x speedup over the Kubernet
 es default scheduler in several cases.\n\nTag: Applications, Emerging Tech
 nologies, Reproducibility and Transparency, Software Engineering, System S
 oftware and Runtime Systems, Workflows\n\nRegistration Category: Workshop 
 Reg Pass
END:VEVENT
END:VCALENDAR