A Neural Network Approach to Forecasting Computing-Resource Exhaustion with Workload

Xue, Ke-Xian; Su, Liang; Jia, Yingmin; Cai, Kai-Yuan

doi:10.1109/qsic.2009.48

Cited by 6 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The principal component analysis method was used to reduce the input variables, and the artificial neural network was used for prediction. In [21], by constructing a multilayer back-propagation neural network and taking the workload as input, the results showed that the prediction effect is good and the workload is proved to impact the software of aging. In [22], a gray correlation artificial neural network (GRANN_ARIMA) method was proposed to mix linear and nonlinear models to predict software aging.…”

Section: Software Aging Predictionmentioning

confidence: 99%

Integrating Artificial Bee Colony Algorithm and BP Neural Network for Software Aging Prediction in IoT Environment

Jing

Meng

2019

IEEE Access

View full text Add to dashboard Cite

Software aging is a common phenomenon that exists in systems that require long periods of operation, especially in Internet-of-Things environments. The back propagation (BP) neural network has been adopted widely to predict the trend of software aging. However, the weight and threshold of the BP neural network are randomly initialized, so it is easy to get the unsatisfactory local optimal solutions and the convergence speed of computing is slow. In this paper, we propose a novel software aging prediction method using the artificial bee colony algorithm to optimize the BP neural network model for achieving better software aging prediction accuracy. The experiment results show that our method fits the prediction trend of software aging more accurately than the traditional BP neural network, and our method also has faster convergence speed and more stable prediction results. INDEX TERMS Artificial bee colony algorithm, BP neural network, software aging, prediction accuracy.

show abstract

Section: Software Aging Predictionmentioning

confidence: 99%

Integrating Artificial Bee Colony Algorithm and BP Neural Network for Software Aging Prediction in IoT Environment

Jing

Meng

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…We used one of these segments in our study, which contains a total of 5,811 observations. Since the 1 A preliminary version of this approach was proposed in QSIC 2009 [33].…”

Section: On the Relationship Between System Workload And Availablmentioning

confidence: 99%

Using Neural Networks to Forecast Available System Resources: An Approach and Empirical Investigation

Jia

Zhi

Xue³

et al. 2015

Int. J. Soft. Eng. Knowl. Eng.

Self Cite

View full text Add to dashboard Cite

Software aging refers to the phenomenon that software systems show progressive performance degradation or a sudden crash after longtime execution. It has been reported that this phenomenon is closely related to the exhaustion of system resources. This paper quantitatively studies available system resources under the realworld situation where workload changes dynamically over time. We propose a neural network approach to first investigate the relationship between available system resources and system workload and then to forecast future available system resources. Experimental results on data sets collected from real-world computer systems demonstrate that the proposed approach is effective.

show abstract

“…The analytical model-based approach uses probability models to estimate the software aging process and determine the optimal rejuvenation scheduling based on the estimated aging state [1], [10]- [12]. On the other hand, measurement-based approaches collect information directly from the system and estimates using statistical or machine learning methods the real state of the system to trigger the rejuvenation process accordingly [13]- [23]. Meanwhile, the former approach is relatively easy to generalize across different systems because it is based on a simplification of the system.…”

Section: Introductionmentioning

confidence: 99%

Optimization of Two-Granularity Software Rejuvenation Policy Based on the Markov Regenerative Process

et al. 2016

Self Cite

View full text Add to dashboard Cite

. Abstract-Software rejuvenation is a proactive software control technique that is used to improve a computing system performance when it suffers from software aging. In this paper, a two-granularity inspection-based software rejuvenation policy, which works as a closed-loop control technique, is proposed. This policy mitigates the negative impact of two-level software aging. The two levels considered are the user-level applications and the operating system. A Markov regenerative process model is constructed based on the system condition. We obtain the degradation rate of the application software and operating system from fault injection experiments. The diagnostic accuracy of the adopted monitor and analysis system, which is applied to inspect the application software and operating system, is considered as we provide the optimal rejuvenation strategies. Finally, the availability and the overall loss probability with their corresponding optimal inspection time intervals are obtained numerically based on the parameter values estimated from the experiments. Experimental results show that two-granularity software rejuvenation is much more effective than traditional single-level software rejuvenation. In our experimental study, when two-granularity software rejuvenation is used, the unavailability and the overall loss probability of the system were reduced by 17.9% and 2.65%, respectively, in comparison with the single-level rejuvenation.Index Terms-Diagnostic accuracy, Markov regenerative process (MRGP), multigranularity software aging, overall loss probability, software rejuvenation.

show abstract

A Neural Network Approach to Forecasting Computing-Resource Exhaustion with Workload

Cited by 6 publications

References 19 publications

Integrating Artificial Bee Colony Algorithm and BP Neural Network for Software Aging Prediction in IoT Environment

Integrating Artificial Bee Colony Algorithm and BP Neural Network for Software Aging Prediction in IoT Environment

Using Neural Networks to Forecast Available System Resources: An Approach and Empirical Investigation

Optimization of Two-Granularity Software Rejuvenation Policy Based on the Markov Regenerative Process

Contact Info

Product

Resources

About