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:20211207T054814Z
LOCATION:220-221
DTSTART;TZID=America/Chicago:20211118T160000
DTEND;TZID=America/Chicago:20211118T163000
UID:submissions.supercomputing.org_SC21_sess173_pap376@linklings.com
SUMMARY:Resilient Error-Bounded Lossy Compressor for Data Transfer
DESCRIPTION:Paper\n\nResilient Error-Bounded Lossy Compressor for Data Tra
 nsfer\n\nLi, Di, Zhao, Liang, Chen...\n\nToday's exascale scientific appli
 cations or advanced instruments are producing vast volumes of data, which 
 need to be shared/transferred through the network/devices with relatively 
 low bandwidth (e.g., WAN). Lossy compression is an important strategy to r
 esolve the big data issue, however, little work was done to make it resili
 ent against silent errors, which may happen during the compression or data
  transferring. In this paper, we propose a resilient error-bounded lossy c
 ompressor. We design a new independent-block-wise model that decomposes ea
 ch dataset into many independent sub-blocks to compress. Then, we design a
 nd implement a series of error detection/correction strategies for each st
 age of SZ. Our solution is the first algorithm-based fault tolerance (ABFT
 ) algorithm for lossy compression. Our solution incurs negligible executio
 n overhead in the fault-free situation. Should soft errors occur, it ensur
 es decompressed data is strictly bounded within user's requirement, with a
  very limited degradation of compression ratio and low overhead.\n\nTag: A
 lgorithms, Cloud and Distributed Computing, Data Management, Parallel Prog
 ramming Languages and Models\n\nRegistration Category: Tech Program Reg Pa
 ss
END:VEVENT
END:VCALENDAR