Fault-Tolerant Multiprocessor Systems Reliability Estimation Using Statistical Experiments with GL-Models

Romankevich, A. M.; Feseniuk, Andrii; Maidaniuk, Ivan; Romankevich, Vitaliy

doi:10.1007/978-3-319-91008-6_19

Cited by 7 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their advantage is the ability to obtain results with high accuracy, but their disadvantage is non-universality: for each type of system, it is often necessary to create a new calculation method [23,24], [25,26], [27,28], [29,30], [31,32], [33,34], [35,36], [37,38], [39,40], [41,42], [43,44], [45,46], [47,48], [49,50], [51,52], [53,54]. The methods of the second group are based on conducting statistical experiments with models of system behavior in the failure flow [55,56], [57]. Their advantage is that they can be used for any type of system, but their disadvantage is that parameter estimation can usually be performed only with a certain level of accuracy, which generally depends on the number of experiments performed.…”

Section: Introductionmentioning

confidence: 99%

On the modification of GL-models by adding edges to a cyclic graph

Romankevich,

Morozov,

Romankevich

et al. 2024

HAIT

View full text Add to dashboard Cite

This work suggests a method for constructing GL-models of fault-tolerant multiprocessor systems. These models can be used, in particular, to estimate the reliability parameters of the latter by conducting statistical experiments with models of their behavior in the failure flow.Two cases are considered: the non-basicsystem, unlike the basicsystem, is resistant to somefailures of increased multiplicity, or else, on the contrary, the non-basicsystem is vulnerableto certain failures that do not lead to the failure of the basicsystem. In this case, the condition under which the system’s behavior differs from the baselinecorresponds to a Boolean expression, thatdepends on the values of the elements of the system state vector, which characterizes the states of its processors in the failure flow. According to the method proposed in the article, amodel of such system is built by adding an edge or several edges to the so-called MLE-model, atype of GL-model, thatcan be constructed for any basicsystem and isbased on cyclic graphs. The edge function for this edge is formed based on the aforementioned Boolean expression. The models constructed by the proposed method are also based on cyclic graphs, which, in particular, significantly simplifythe procedure for assessing the connectivityof the last ones. A series of experiments have been conducted to confirm the adequacy of the models (obtained by the proposed method) to the behavior of systems in the failure flow. This work presents examplesthat demonstratethe process of constructing GL-models for non-basicfault-tolerant multiprocessor systemsusingthe proposed method for both of the above cases.

show abstract

Section: Introductionmentioning

confidence: 99%

On the modification of GL-models by adding edges to a cyclic graph

Romankevich,

Morozov,

Romankevich

et al. 2024

HAIT

View full text Add to dashboard Cite

show abstract

“…The methods of the second group are based on statistical experiments with models of the behavior of the fault-tolerant multiprocessor systems (FTMS) in the failure flow [55,56]. Their advantage is their flexibility, while the disadvantage is the need to conduct a large number of experiments with models.…”

Section: Introductionmentioning

confidence: 99%

“…The building of non-basic GL-models (i.e., models of non-basic systems) can be performed by modifying basic ones. This modification can occur by changing the structure of the graph [62], by modifying the expressions of the edge functions [63], or by combining both approaches.…”

Section: Introductionmentioning

confidence: 99%

On the method of building of non-basic GL-models which are formed on combination of edge functions of basic models

Romankevich,

Morozov,

Romankevich

et al. 2024

AAIT

View full text Add to dashboard Cite

Thiswork is dedicatedto the problem of building GL-models of behaviorof non-basicfault-tolerant multiprocessor systems in the failure flow. Such models can be used to calculate the reliability parameters of the latter. The system, depending on the fulfillment of certain conditions, is resistant to failures of various multiplicities. These conditions depend only on the states of the system's processors and can be represented by special Boolean expressions. A method ofconstructing GL-models of such systems based on combining expressions of the edge functions of auxiliary basicmodelsis proposed. At the same time, the specific featureof the models that were built by the proposed methodis that they are based on cyclic graphs. This simplifies the process of evaluating their connectivity and also simplifies the analysis of the model's operation. In addition, this allows usingother methods of modifying GL-models if necessary. The method involves usingauxiliary models that have the same number of edges.In order toequalize the number of edges, auxiliary models can be extendedwith additional edges with edge functions of a special type. It is shown that this extensiondoes not change the behavior of the models. In particular, theprocedure oforthogonalization ofthe Boolean expressions is described, which should be carried out if the conditions can be satisfiedsimultaneously. It is shown that the expressions of the edge functions of the obtained GL-models, which can be quite complex, can sometimes be significantly simplified. Numerous experiments have been conducted to confirmthe adequacy of the models (builtby the proposed method) to the behavior of the corresponding systems in the failure flow. Theexample is given to demonstratethe application of the proposed method.The resulting model is analyzed and shown to correspond to the behavior of the system forwhich it was built.

show abstract