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:20260522T150115Z
LOCATION:C141/143/149
DTSTART;TZID=America/Chicago:20181115T133000
DTEND;TZID=America/Chicago:20181115T140000
UID:submissions.supercomputing.org_SC18_sess192_pap247@linklings.com
SUMMARY:Optimizing Software-Directed Instruction Replication for GPU Error
  Detection
DESCRIPTION:Abdulrahman Mahmoud (University of Illinois) and Siva Kumar Sa
 stry Hari, Michael B. Sullivan, Timothy Tsai, and Stephen W. Keckler (Nvid
 ia Corporation)\n\nApplication execution on safety-critical and high-perfo
 rmance computer systems must be resilient to transient errors. As GPUs bec
 ome more pervasive in such systems, they must supplement ECC/parity for ma
 jor storage structures with reliability techniques that cover more of the 
 GPU hardware logic.  Instruction duplication has been explored for CPU res
 ilience; however, it has never been studied in the context of GPUs, and it
  is unclear whether the performance and design choices it presents makes i
 t a feasible GPU solution. This paper describes a practical methodology to
  employ instruction duplication for GPUs and identifies implementation cha
 llenges that can incur high overheads (69% on average). It explores GPU-sp
 ecific software optimizations that trade fine-grained recoverability for p
 erformance. It also proposes simple ISA extensions with limited hardware c
 hanges and area costs to further improve performance, cutting the runtime 
 overheads by more than half to an average of 30%.\n\nTag: Algorithms, Arch
 itectures, GPUs, Linear Algebra, Networks, Resiliency\n\nRegistration Cate
 gory: Tech Program Reg Pass\n\nSession Chair: Steven A. Wright (University
  of York, England)\n\n
END:VEVENT
END:VCALENDAR
