S-Shaped Reliability Growth Modeling for Software Error Detection

Yamada, Shigeru; Ohba, Mitsuru; Osaki, Shunji

doi:10.1109/tr.1983.5221735

Cited by 733 publications

(284 citation statements)

References 12 publications

Supporting

Mentioning

270

Contrasting

Unclassified

Order By: Relevance

“…) λyex(t) = abre −bt e −r(1−e −bt ) Gompertz [7], [9], [11] GOMP S-shaped We have compared eight most widely used NHPP models for modelling of the fault detection process: MusaLogarithmic, Goel-Okumoto Exponential, Generalized GoelOkumoto, Inflection S-shaped, Delayed S-Shaped, YamadaExponential, Gompertz and Logistic, whose mean value function and failure intensity are given in the Table I.…”

Section: A Fault Detection Process As Nhppmentioning

confidence: 99%

An empirical study of software reliability in SDN controllers

Vizarreta

Trivedi

Helvik

et al. 2017

2017 13th International Conference on Network and Service Management (CNSM)

View full text Add to dashboard Cite

Abstract-Software Defined Networking (SDN) exposes critical networking decisions, such as traffic routing or enforcement of the critical security policies, to a software entity known as the SDN controller. Controller software, as written by humans, is intrinsically prone to bugs, which may impair the network performance as a whole, if activated. Software reliability growth models (SRGM) are often used to estimate and predict the reliability of the software in the operational phase based on the fault report data during the testing phase. These models can be used to predict the number of residual bugs in the software, as well as failure intensity, software reliability and optimal software release time. In this paper we analyze ten releases of ONOS open source controller, whose uncensored fault reports are available online.

show abstract

Section: A Fault Detection Process As Nhppmentioning

confidence: 99%

An empirical study of software reliability in SDN controllers

Vizarreta

Trivedi

Helvik

et al. 2017

2017 13th International Conference on Network and Service Management (CNSM)

View full text Add to dashboard Cite

show abstract

“…The following [11] a=135.965, α =.138, η=1.000E-4 2TL1 [5] a=135.965, α =49.216, η1 =.003, η2 =1.000E-4 2TL2 [6] a=135.974, α =2.611, η1 =2.566, η2 =.001, τ =6.562 Comparison under dataset [2] Model Parameters G-O [1] a= 218.159, b=.041 Chiu [11] a=215.706, α =.042, η=.001 2TL1 [5] a=215.706, α =56.69, η1 =.001, η2 =.001 2TL2 [6] a=210.134, α =.094, η1 =.442, η2 =1.000E-5, τ =7.007E-5 Comparison under dataset [3] Model Parameters G-O [1] a= 33.6, b=.063 Chiu [11] a=24.821, α =.024, η=.343 2TL1 [5] a=24.821, α =.056, η1 =.424, η2 =.343 2TL2 [6] a=24.821, α =.217, η1 =.658, η2 =.343, τ =.119 Comparison under dataset [4] Model Parameters G-O [1] a= 133.761, b=.015 Chiu [11] a=133.496, α =.146, η=.001 2TL1 [5] a=133.496, α =153.843, η1 =.001, η2 =.001 2TL2 [6] a=133.496, α =.002, η1 =909.569, η2 =.001, τ =1.748 Comparison under dataset [5] Model Parameters G-O [1] a= 18.257, b=.397 Chiu [11] a=18.254, α =.397, η=.001 2TL1 [5] a=18.254, α =.049, η1 =8.151, η2 =.001 2TL2 [6] a=18.257, α =23.658, η1 =.665, η2 =1.000E-5, τ =15.341 Comparison under dataset [6] Model Parameters G-O [1] a= 124.44, b=.051 Chiu [11] a=124.171, α =.051, η=.001 2TL1 [5] a=124.171, α =.001, η1 =88.486, η2 =.001 2TL2 [6] a=124.437, α =.163, η1 =5.874, η2 =1.000E-5, τ =.909 Comparison under dataset [7] Model Parameters G-O [1] a= 23.092, b=.559 Chiu [11] a=22.252, α =.493, η=.332 2TL1 [5] a=22.252, α =.211, η1 =.2.338, η2 =.332 2TL2 [6] a=22.252, α =9.048, η1 =.195, η2 =.332, τ =1.272…”

Section: Parameter Estimationmentioning

confidence: 99%

“…Goel and Okumoto in [1] proposed an exponential SRGM. Yamada and Ohba in [2] proposed delayed S-shaped SRGM while Ohba in [3] proposed inflection S-shaped SRGM. Gokhale and Trivedi in [4] proposed an enhanced NHPP model which takes into account the time-dependent failures occurring in debugging process.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Some Software Reliability Growth Models with Learning Effects

Iqbal¹

2016

IJMSC

View full text Add to dashboard Cite

A newly developed software system before its deployment is subjected to vigorous testing so as to minimize the probability of occurrence of failure very soon. Software solutions for safety critical and mission-critical application areas need a much focused level of testing. The testing process is basically carried out to build confidence in the software for its use in real world applications. Thus, reliability of systems is always a matter of concern for us. As we keep on performing the error detection and correction process on our software, the reliability of the system grows. In order to model this growth in the system reliability, many formulations in Software Reliability Growth Models (SRGMs) have been proposed including some based on NonHomogeneous Poisson Process (NHPP). The role of human learning and experiential pattern gains are being studied and incorporated in such models. The realistic assumptions about human learning behavior and experiential gains of new skill-sets for better detection and correction of faults on software are being incorporated and studied in such models. In this paper, a detailed analysis of some select SRGMs with learning effects is presented based on use of seven data sets. The estimation of parameters and comparative analysis based on goodness of fit using seven data sets are presented. Moreover, model comparisons on the basis of total defects predicted by the select models are also tabulated.Index Terms: Software Reliability, Software Reliability Growth Model (SRGM), Non-Homogeneous Poisson Process (NHPP), Learning effect, two-type learning effect.

show abstract

“…В основу більшості таких моделей покладено розподіл Пуассона, параметри якого мають різний вигляд для різних мо-делей. Найбільш поширеними моделями цього класу є моделі Гоеля-Окумото [11], S-подібна модель зростання надійності Ямади [12] тощо. Проте, основним недоліком відомих моделей на основі пуассонового процесу є те, що внаслідок припущень і спрощень, вони не достатньо адекватно відображають процес тестування, а результа-ти, отримані при їх застосуванні, не завжди збігаються з отриманими на практиці [13].…”

Section: аналіз літературних даних та постановка проблемиunclassified