2008
DOI: 10.1109/l-ca.2007.19
|View full text |Cite
|
Sign up to set email alerts
|

Computing Accurate AVFs using ACE Analysis on Performance Models: A Rebuttal

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
17
0

Year Published

2008
2008
2021
2021

Publication Types

Select...
5
2
2

Relationship

0
9

Authors

Journals

citations
Cited by 37 publications
(17 citation statements)
references
References 4 publications
0
17
0
Order By: Relevance
“…Wang et al show that lack of model detail in ACE Analysis can lead to overly conservative AVF estimates [10]. This was shown to not be a fundamental limitation of the technique, however, and the authors themselves conclude that AVF Analysis is an indispensable tool for early-design reliability analysis [11]. Li et al show that FIT rates calculated using average AVF values may be inaccurate since AVF values are often not independent [12].…”
Section: Architectural Vulnerability Factormentioning
confidence: 99%
“…Wang et al show that lack of model detail in ACE Analysis can lead to overly conservative AVF estimates [10]. This was shown to not be a fundamental limitation of the technique, however, and the authors themselves conclude that AVF Analysis is an indispensable tool for early-design reliability analysis [11]. Li et al show that FIT rates calculated using average AVF values may be inaccurate since AVF values are often not independent [12].…”
Section: Architectural Vulnerability Factormentioning
confidence: 99%
“…Their results corroborated that ACE analysis overestimates microprocessor vulnerability, mostly due to less detailed structures available in the performance model employed in [15].Furthermore, recent radiation studies with actual proton and neutron irradiation tests showed that SFI measurements closely match in-field exposure [18]. While [19] shows that ACE inaccuracy can be reduced by adding extra detail to the employed model, accurate AVF analysis heavily relies on the existence of a detailed RTL model.…”
Section: Related Workmentioning
confidence: 69%
“…Physical interleaving, on the other hand, assigns physically-adjacent bits to different data words, thus ensuring adjacent bits are protected by different [41]. This analysis was rebutted, and the effects pointed out by Wang were incorporated into an ACE analysis infrastructure [8]. Sridharan and Kaeli extended the notion of vulnerability to allow independent measurement of the vulnerability of programs and hardware [32] [33].…”
Section: Existing Multi-bit Remediation Techniquesmentioning
confidence: 99%