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:20211207T054812Z
LOCATION:230-231-232
DTSTART;TZID=America/Chicago:20211118T133000
DTEND;TZID=America/Chicago:20211118T140000
UID:submissions.supercomputing.org_SC21_sess169_pap312@linklings.com
SUMMARY:Error-Controlled, Progressive, and Adaptable Retrieval of Scientif
 ic Data with Multilevel Decomposition
DESCRIPTION:Paper\n\nError-Controlled, Progressive, and Adaptable Retrieva
 l of Scientific Data with Multilevel Decomposition\n\nLiang, Gong, Chen, W
 hitney, Wan...\n\nExtreme-scale simulations and high-resolution instrument
 s are generating an increasing amount of data, which poses significant cha
 llenges to both data storage and retrieval. The challenges in satisfying v
 arious analysis needs while minimizing I/O overhead should never be left u
 nmanaged. In this paper, we propose a data refactoring/compressing/retriev
 al framework capable of: fine-grained data refactoring with regard to prec
 ision; incremental retrieving and recomposing data toward requested error 
 bounds; and adaptively retrieving data in multi-precision and multi-resolu
 tion with respect to analysis. Our framework reduces the amount of data re
 trieved when multiple incremental precisions are requested. Experiments sh
 ow that the amount of data retrieved under the same progressively requeste
 d distortion using our method is 64% less than that using state-of-the-art
  approaches. Parallel experiments with up to 1024 cores and ~600GB data sh
 ow that our approach yields 1.36x and 2.52x performance over existing appr
 oaches in writing to and reading from persistent storage systems, respecti
 vely.\n\nTag: Big Data, Data Analytics, Data Management, File Systems and 
 I/O, Memory Systems, Storage\n\nRegistration Category: Tech Program Reg Pa
 ss
END:VEVENT
END:VCALENDAR