Abstract

We propose a processor-level technique called Selective Replication, by which the application can choose where in its application stream and to what degree it requires replication. Recent work on static analysis and fault-injection-based experiments on applications reveals that certain variables in the application are critical to its crash- and hang-free execution. If it can be ensured that only the computation of these variables is error-free, then a high degree of crash/hang coverage can be achieved at a low performance overhead to the application. The Selective Replication technique provides an ideal platform for validating this claim. The technique is compared against complete duplication as provided in current architecture-level techniques. The results show that with about 59% less overhead than full duplication, selective replication detects 97% of the data errors and 87% of the instruction errors that were covered by full duplication. It also reduces the detection of errors benign to the final outcome of the application by 17.8% as compared to full duplication.

Original languageEnglish (US)
Title of host publicationProceedings - 37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2007
Pages544-553
Number of pages10
DOIs
StatePublished - Nov 16 2007
Event37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2007 - Edinburgh, United Kingdom
Duration: Jun 25 2007Jun 28 2007

Publication series

NameProceedings of the International Conference on Dependable Systems and Networks

Other

Other37th Annual IEEE/IFIP International Conference on Dependable Systems and Networks, DSN 2007
Country/TerritoryUnited Kingdom
CityEdinburgh
Period6/25/076/28/07

Keywords

  • Application-aware
  • Critical variable
  • Duplication
  • Error detection
  • Redundant hardware

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Processor-level selective replication'. Together they form a unique fingerprint.

Cite this