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:20181221T160726Z
LOCATION:D220
DTSTART;TZID=America/Chicago:20181111T144500
DTEND;TZID=America/Chicago:20181111T144800
UID:submissions.supercomputing.org_SC18_sess160_ws_whpc127@linklings.com
SUMMARY:Characterization of the Impact of Soft Errors on Iterative Methods
DESCRIPTION:Workshop\nDiversity, Education, Hot Topics, Workshop Reg Pass\
 n\nCharacterization of the Impact of Soft Errors on Iterative Methods\n\nM
 utlu, Kestor, Unsal, Krishnamoorthy\n\nSoft errors caused by transient bit
  flips can lead to silent data corruption which can significantly impact a
 n application’s behavior. When a transient bit flip affects a hardware com
 ponent, the application is said to be impacted by a soft error. When a sof
 t error escapes hardware detection and impacts the application state, it c
 an impact execution by leading to incorrect results or significantly impac
 ting application execution times. Recent architectural trends, such as nea
 r-threshold voltage operation and constrained power budgets, exacerbate th
 e frequency and impact of soft errors. This has motivated design of numero
 us soft error detection and correction techniques, which focuses localizin
 g the application vulnerabilities, and then correcting these errors using 
 micro-architectural, architectural, compilation-based, or application-leve
 l techniques. \n\nA broad array of techniques has been designed to underst
 and application behavior under soft errors and to detect, isolate, and cor
 rect soft-error-impacted application state. The first step toward tolerati
 ng soft errors involves understanding an application’s behavior under soft
  errors. This can help elucidate the need for error detection/correction t
 echniques.\n\nIn this study, we designed a deterministic application level
  error injection method that allows us explore the error injection space o
 ver several iterative solvers. As iterative methods are a crucial componen
 t of scientific applications, and consume a significant fraction of superc
 omputing time, we performed this study on the iterative methods. We used r
 eal life datasets to evaluate and characterize vulnerabilities and strengt
 hs of each solver under identical settings.
URL:https://sc18.supercomputing.org/presentation/?id=ws_whpc127&sess=sess1
 60
END:VEVENT
END:VCALENDAR