BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20241120T082409Z
LOCATION:HG E 1.1
DTSTART;TZID=Europe/Stockholm:20240605T120000
DTEND;TZID=Europe/Stockholm:20240605T123000
UID:submissions.pasc-conference.org_PASC24_sess121_msa263@linklings.com
SUMMARY:Challenges and Opportunities in Running Kubernetes Workloads on HP
 C
DESCRIPTION:Minisymposium\n\nAntony Chazapis (FORTH); Fotis Nikolaidis (Su
 perDuperDB); Manolis Marazakis (FORTH); and Angelos Bilas (FORTH, Universi
 ty of Crete)\n\nCloud and HPC increasingly converge in hardware platform c
 apabilities and specifications, \nnevertheless still largely differ in the
  software stack and how it manages available resources.\nThe HPC world typ
 ically favors Slurm for job scheduling, whereas Cloud deployments rely on 
 \nKubernetes to orchestrate container instances across nodes. Running hybr
 id workloads is possible by using bridging mechanisms that submit jobs fro
 m one environment to the other.\nHowever, such solutions require costly da
 ta movements, while operating within the constraints \nset by each setup's
  network and access policies. In this presentation, we introduce an contai
 ner-based approach design that enables running unmodified Kubernetes workl
 oads directly on HPC systems, by having users deploy their own private Kub
 ernetes mini Cloud, which internally converts container lifecycle manageme
 nt commands to use the HPC system-level Slurm infrastructure for schedulin
 g and Singularity/Apptainer as the container runtime. We consider this app
 roach to be practical for deployment in HPC centers, as it requires minima
 l pre-configuration and retains existing resource management and accountin
 g policies.\n\nDomain: Climate, Weather, and Earth Sciences, Engineering, 
 Computational Methods and Applied Mathematics\n\nSession Chair: Tiziano Mü
 ller (HPE)
END:VEVENT
END:VCALENDAR
