BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160031Z
LOCATION:Track 8
DTSTART;TZID=America/New_York:20201118T163500
DTEND;TZID=America/New_York:20201118T164600
UID:submissions.supercomputing.org_SC20_sess355_spostu117@linklings.com
SUMMARY:Recovering Silent Data Corruption through Spatial Prediction
DESCRIPTION:ACM Student Research Competition: Graduate Poster, ACM Student
  Research Competition: Undergraduate Poster\n\nRecovering Silent Data Corr
 uption through Spatial Prediction\n\nPlacke\n\nHigh-performance computing 
 applications are central to advancement in many fields of science and engi
 neering. Central to this advancement is the supposed reliability of the HP
 C system. However, as system size grows and hardware components are run wi
 th near-threshold voltages, transient upset events become more likely. Man
 y works have explored the problem of detection of silent data corruption. 
 Recovery is often left to checkpoint-restart or application-specific techn
 iques. This poster explores the use of spatial similarity to recover from 
 silent data corruption. We explore eight reconstruction methods and find t
 hat Linear Regression yields the best results with over 90% of Linear Regr
 ession’s corrections having less than 1% relative error.\n\nTag: Student P
 rogram\n\nRegistration Category: Tech Program Reg Pass
END:VEVENT
END:VCALENDAR