Proceedings 16th International Parallel and Distributed Processing Symposium 2002
DOI: 10.1109/ipdps.2002.1016526
|View full text |Cite
|
Sign up to set email alerts
|

Optimizing the reliability of component-based n-version approaches

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
12
0

Year Published

2007
2007
2017
2017

Publication Types

Select...
3
2
1

Relationship

0
6

Authors

Journals

citations
Cited by 8 publications
(12 citation statements)
references
References 4 publications
0
12
0
Order By: Relevance
“…The selection problem is formulated as a mixed integer programming problem, and some heuristics are proposed to calculate in an efficient way an approximate solution. An analogous problem is considered in [38]. The proposed methodology is motivated by its application to componentbased systems, but it can be easily extended to a SOA environment.…”
Section: Related Workmentioning
confidence: 99%
“…The selection problem is formulated as a mixed integer programming problem, and some heuristics are proposed to calculate in an efficient way an approximate solution. An analogous problem is considered in [38]. The proposed methodology is motivated by its application to componentbased systems, but it can be easily extended to a SOA environment.…”
Section: Related Workmentioning
confidence: 99%
“…The forgetting element is necessary to ensure the operative response of the system to changes in the behavior of versions when a certain version can significantly change its reliability in case of changing the input data stream [3], therefore is necessary to change rapidly its rating of version for correct distribution of the weights during voting [4]. The rating is determined by summing of all the elements of the stack.…”
Section: Voting Algorithmsmentioning
confidence: 99%
“…Each selected concrete service is then dynamically bound to the corresponding abstract service. Other methodologies [4,8,9] extend this idea by also considering the possibility of binding each abstract service to a set of functionally equivalent concrete services rather than a single service, coordinated according to some redundancy pattern (e.g., one-out-of-N or sequential retry), to achieve higher QoS at the expense of higher cost: in this case both the redundancy pattern and the set of equivalent concrete services must be selected at runtime.…”
Section: Introductionmentioning
confidence: 99%