2016
DOI: 10.1109/tpds.2015.2426179
|View full text |Cite
|
Sign up to set email alerts
|

A Survey of Techniques for Modeling and Improving Reliability of Computing Systems

Abstract: Recent trends of aggressive technology scaling have greatly exacerbated the occurrences and impact of faults in computing systems. This has made 'reliability' a first-order design constraint. To address the challenges of reliability, several techniques have been proposed. This paper provides a survey of architectural techniques for improving resilience of computing systems. We especially focus on techniques proposed for microarchitectural components, such as processor registers, functional units, cache and mai… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
44
0

Year Published

2016
2016
2023
2023

Publication Types

Select...
7
1
1

Relationship

3
6

Authors

Journals

citations
Cited by 71 publications
(45 citation statements)
references
References 114 publications
1
44
0
Order By: Relevance
“…However, through WDEs, a particular cell can be attacked by writing to many of its neighbors. Also, crossing of the WRE limit due to repeated writes leads to hard errors (i.e, failure of the memory cell) in NVMs, whereas WDE leads to soft errors (i.e., flipping of the bit) [47]. Since flipping of the stored value is likely to get unnoticed more easily than the failure of the memory, WDE-induced attacks are more difficult to detect.…”
Section: Conclusion and Future Outlookmentioning
confidence: 99%
“…However, through WDEs, a particular cell can be attacked by writing to many of its neighbors. Also, crossing of the WRE limit due to repeated writes leads to hard errors (i.e, failure of the memory cell) in NVMs, whereas WDE leads to soft errors (i.e., flipping of the bit) [47]. Since flipping of the stored value is likely to get unnoticed more easily than the failure of the memory, WDE-induced attacks are more difficult to detect.…”
Section: Conclusion and Future Outlookmentioning
confidence: 99%
“…With increasing core count, the cache and main memory capacity is on the rise [11], and due to this, the overhead of correction techniques, such as scrubbing, is also growing [12]. STT-RAM and PCM are used for cache and main memory where any additional latency leads to a large impact on system performance.…”
Section: Need For Improving Nvm Reliabilitymentioning
confidence: 99%
“…Focusing on hardware engineering, Sparsh M. et al provided a comprehensive survey of reliability techniques for microarchitecture design (such as processor registers, cache and main memory, etc.) [152]. Thaha M. et al surveyed the fault detection algorithms in WSN and performed a qualitative comparison of the latest fault detection researches including centralized, distributed and hybrid algorithms and analyzed the shortcomings, advantages of those algorithms.…”
Section: The Traditional Dependability Means and Vandvmentioning
confidence: 99%