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:20211207T055413Z
LOCATION:222
DTSTART;TZID=America/Chicago:20211119T105000
DTEND;TZID=America/Chicago:20211119T111000
UID:submissions.supercomputing.org_SC21_sess331_ws_scc101@linklings.com
SUMMARY:Multi-Tenancy Management and Zero Downtime Upgrades Using Cray-HPE
  Shasta Supercomputers
DESCRIPTION:Workshop\n\nMulti-Tenancy Management and Zero Downtime Upgrade
 s Using Cray-HPE Shasta Supercomputers\n\nAlam, Gila, Klein, Martinasso\n\
 nAs supercomputing systems gradually become an integral part of data drive
 n workflows such as ML/AI or tightly-coupled pre/post-processing pipelines
 , users need programmable access to shared resources to avoid moving large
  volume of data to dedicated systems or to clouds.  Public clouds or the p
 rivate ones using technologies like OpenStack, multi-tenancy on shared har
 dware is a commonplace, offering users programmable and privileged access 
 to compute, network and storage resources. Such access is unavailable to u
 sers on batch-scheduled, multi-Petascale supercomputing systems, which are
  designed for achieving  close-to-metal performance for scientific applica
 tions at unprecedented scales.  In this paper, we focus on multi-tenancy w
 ithin hardware and software stacks of Cray-HPE EX Shasta supercomputing sy
 stems for creating high performance and cloud virtual-clusters for HPC and
  AI/ML workloads respectively. Using orchestration examples for zero-downt
 ime upgrades of virtual-clusters, we demonstrate benefits of multi-tenancy
  for achieving close-to-metal performance,  alongside resource elasticity 
 and customization, without interruption to operational services.\n\nTag: C
 loud and Distributed Computing, Interactive HPC, Productivity Tools\n\nReg
 istration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR