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:20260522T150117Z
LOCATION:C141/143/149
DTSTART;TZID=America/Chicago:20181115T160000
DTEND;TZID=America/Chicago:20181115T163000
UID:submissions.supercomputing.org_SC18_sess186_pap521@linklings.com
SUMMARY:Stacker: An Autonomic Data Movement Engine for Extreme-Scale Data 
 Staging-Based In Situ Workflows
DESCRIPTION:Pradeep Subedi, Philip Davis, and Shaohua Duan (Rutgers Univer
 sity); Scott Klasky (Oak Ridge National Laboratory); Hemanth Kolla (Sandia
  National Laboratories); and Manish Parashar (Rutgers University)\n\nData 
 staging and in situ workflows are being explored extensively as an approac
 h to address data-related costs at very large scales. However, the impact 
 of emerging storage architectures (e.g., deep memory hierarchies and burst
  buffers) upon data staging solutions remains a challenge. In this paper, 
 we investigate how burst buffers can be effectively used by data staging s
 olutions, for example, as a persistence storage tier of the memory hierarc
 hy. Furthermore, we use machine learning based prefetching techniques to m
 ove data between the storage levels in an autonomous manner. We also prese
 nt Stacker, a prototype of the proposed solutions implemented within the D
 ata\-Spaces data staging service, and experimentally evaluate its performa
 nce and scalability using the S3D combustion workflow on current leadershi
 p class platforms. Our experiments demonstrate that Stacker achieves low l
 atency, high volume data-staging with low overhead as compared to in-memor
 y staging services for production scientific workflows.\n\nTag: Architectu
 res, Data Management, File Systems, Networks, State of the Practice, Syste
 m Software, Workflows\n\nRegistration Category: Tech Program Reg Pass\n\nS
 ession Chair: Robin J. Goldstone (Lawrence Livermore National Laboratory)\
 n\n
END:VEVENT
END:VCALENDAR