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:20210916T132450Z
LOCATION:Mère Royaume
DTSTART;TZID=Europe/Stockholm:20210707T110000
DTEND;TZID=Europe/Stockholm:20210707T113000
UID:submissions.pasc-conference.org_PASC21_sess154_msa114@linklings.com
SUMMARY:Scalable Data-Driven Workflow in SELVEDAS
DESCRIPTION:Minisymposium\n\nScalable Data-Driven Workflow in SELVEDAS\n\n
 Leong, Alam\n\nSupercomputing ecosystems optimize cost and scaling for com
 puting and storage resources by exploiting typically a shared batch access
  model, which is optimized for high utilization of compute resources.  In 
 comparison, in public clouds, on-demand service delivery models address th
 e concept of elasticity while maintaining isolation with performance trade
 -offs. These on-demand access models allow for different degrees of privil
 eges to users for managing IT infrastructure services, in contrast with sh
 ared, bare-metal supercomputing ecosystems. We present an approach for ena
 bling interactive, on-demand supercomputing for experimental data-driven w
 orkflows. These experimental facility workflows are characterised by manag
 ed but bursty data and computing requirements that reflect the workflows. 
 Offline post-processing of archived data in petabyte scale advances both p
 erformance and scalability needs, and depends on advanced data and network
  solutions. The workflow specification also encompasses undertaking the pr
 ovision of a secured data sharing model to facilitate the distribution of 
 processed data with non-registered users. A delegated batch reservation mo
 del is also introduced, controlled by the customer at the experimental fac
 ility and provisioned by the supercomputing site. This allows scientists t
 o couple generation of data to the allocation of compute, data and network
  resources at the supercomputing centre in a transparent manner and  manag
 e resources at both the experimental and supercomputing facilities interac
 tively. Prototype implementation demonstrates that this simple co-designed
  extension to a supercomputing classic batch scheduling system with a cont
 rolled degree of privilege can be easily incorporated to the experimental 
 facilities existing IT resource management and scheduling pipelines.\n\nDo
 main: CS and Math, Emerging Applications, Chemistry and Materials, Physics
 , Life Sciences, Engineering
END:VEVENT
END:VCALENDAR