A workload-based nonlinear approach for predicting available computing resources

doi:10.21629/JSEE.2020.01.21

Journal of Systems Engineering and Electronics ›› 2020, Vol. 31 ›› Issue (1): 224-230.doi: 10.21629/JSEE.2020.01.21

• Reliability • Previous Articles

A workload-based nonlinear approach for predicting available computing resources

Yunfei JIA^1,*(), Zhiquan ZHOU²(), Renbiao WU¹()

¹ Tianjin Key Laboratory for Advanced Signal Processing, Civil Aviation University of China, Tianjin 300300, China
² School of Computing and Information Technology, University of Wollongong, Wollongong 2522, Australia

Received:2019-03-22 Online:2020-02-20 Published:2020-02-25
Contact: Yunfei JIA E-mail:yfjia@cauc.edu.cn;zhiquan@uow.edu.au;rbwu@cauc.edu.cn
About author:JIA Yunfei was born in 1979. He is currently an associate professor at Civil Aviation University of China. He received his B.E. (2001) and M.S. (2004) degrees from Hebei University of Technology, and completed a Ph.D. in software testing at Beihang University in 2010. His research interests include software testing and software reliability modelling. E-mail: yfjia@cauc.edu.cn|ZHOU Zhiquan was born in 1976. He received his B.S. degree in computer science from Peking University, China, and the Ph.D. degree in software engineering from The University of Hong Kong. He is currently an associate professor at the University of Wollongong, Australia. His current research interest is in software testing and analysis. E-mail: zhiquan@uow.edu.au|WU Renbiao was born in 1966. He received his M.S. degree from Northwestern Polytechnical University and completed his Ph.D. degree in signal processing at Xidian University. Currently, he is a professor at Civil Aviation University of China. His interests are signal processing and image processing. E-mail: rbwu@cauc.edu.cn
Supported by:
the Natural Science Foundation of Tianjin(19JCYBJC15900);the National Key Research and Development Program of China(2018YFC0823701);an Open Fund of Tianjin Key Lab for Advanced Signal Processing(2017ASP-TJ04);a linkage grant of the Australian Research Council(LP160101691);This work was supported by the Natural Science Foundation of Tianjin (19JCYBJC15900), the National Key Research and Development Program of China (2018YFC0823701), an Open Fund of Tianjin Key Lab for Advanced Signal Processing (2017ASP-TJ04), and a linkage grant of the Australian Research Council (LP160101691)

Abstract

Abstract:

Performance degradation or system resource exhaustion can be attributed to inadequate computing resources as a result of software aging. In the real world, the workload of a web server varies with time, which will cause a nonlinear aging phenomenon. The nonlinear property often makes analysis and modelling difficult. Workload is one of the important factors influencing the speed of aging. This paper quantitatively analyzes the workload-aging relation and proposes a framework for aging control under varying workloads. In addition, this paper proposes an approach that employs prior information of workloads to accurately forecast incoming system exhaustion. The workload data are used as a threshold to divide the system resource usage data into multiple sections, while in each section the workload data can be treated as a constant. Each section is described by an individual autoregression (AR) model. Compared with other AR models, the proposed approach can forecast the aging process with a higher accuracy.

Key words: software aging, nonlinear phenomenon, fault forecasting

Yunfei JIA, Zhiquan ZHOU, Renbiao WU. A workload-based nonlinear approach for predicting available computing resources[J]. Journal of Systems Engineering and Electronics, 2020, 31(1): 224-230.

Figures/Tables 7

Fig 1

Fig 2

Fig 3

Fig 4

Fig 5

Table 1

Table 2

References 13

1	BRUNEO D, DISTEFANO S, LONGO F, et al. Workload-based software rejuvenation in cloud systems. IEEE Trans. on Computers, 2013, 62 (6): 1072- 1085. doi: 10.1109/TC.2013.30
2	YAN Y Q, GUO P, CHENG B, et al. An experimental case study on the relationship between workload and resource consumption in a commercial web server. Journal of Computational Science, 2018, 25, 183- 192. doi: 10.1016/j.jocs.2017.05.019
3	MATIAS R, BARBETTA P A, TRIVEDI K S, et al. Accelerated degradation tests applied to software aging experiments. IEEE Trans. on Reliability, 2009, 59 (1): 102- 114.
4	GROTTKE M, LI L, VAIDYANATHAN K, et al. Analysis of software aging in a web server. IEEE Trans. on Reliability, 2006, 55 (3): 411- 420. doi: 10.1109/TR.2006.879609
5	VAIDYANATHAN A, TRIVEDI K S. A comprehensive model for software rejuvenation. IEEE Trans. on Dependable and Secure Computing, 2005, 2 (2): 124- 137. doi: 10.1109/TDSC.2005.15
6	CONTRONEO D, NATELLA R, PIETRANTUONO R, et al. A survey of software aging and rejuvenation studies. ACM Journal on Emerging Technologies in Computing Systems, 2014, 10 (1): 1- 34.
7	TIAN J, RUDRARAJU S, LI Z. Evaluating web software reliability based on workload and failure data extracted from server logs. IEEE Trans. on Software Engineering, 2004, 30 (11): 754- 769. doi: 10.1109/TSE.2004.87
8	FUMIO M, XIANG J W, TADANO K, et al. Lifetime extension of software execution subject to aging. IEEE Trans. on Reliability, 2017, 66 (1): 123- 134. doi: 10.1109/TR.2016.2615880
9	MOHAMED E, ZAYNEB T, KAMEL B. Workload-dependent software aging impact on performance and energy consumption in server virtualized systems. Proc. of the IEEE International Symposium on Software Reliability Engineering, 2016: 111-118.
10	ZHAO J, WANG Y B, NING G R, et al. A comprehensive approach to optimal software rejuvenation. Performance Evaluation, 2013, 70 (11): 917- 933. doi: 10.1016/j.peva.2013.05.010
11	SEN P K. Estimates of the regression coefficient based on Kendall's Tau. Journal of the American Statistical Association, 1968, 63 (324): 1379- 1389. doi: 10.1080/01621459.1968.10480934
12	BOX G E P, JENKINS G M, REINSEL G C. Time series analysis: forecasting and control. Hoboken: Prentice-Hall Inc., 1994.
13	DAVIDSON J. Establishing conditions for the functional central limit theorem in nonlinear and semiparametric time series processes. Journal of Econometrics, 2002, 106 (2): 243- 269. doi: 10.1016/S0304-4076(01)00100-2

Available resources	MSE (AR, our approach)	MMRE (AR, our approach)	MBRE (AR, our approach)
memFree	$(9.29\times10^{11}, 7.34\times10^{11})$	(0.048 2, 0.044 4)	(-0.027 7, -0.014 9)
usedSwap	$(9.01\times10^{11}, 7.54\times10^{11})$	(0.009 6, 0.008 7)	(-0.003 8, -0.002 8)

Workload/(req/s)	Exhaustion time for memory/min	Exhaustion time for swap/min
350	16 500	120 000
360	8 900	65 000
390	3 700	45 000

A workload-based nonlinear approach for predicting available computing resources

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

Share this article

Figures/Tables 7

References 13

Related Articles 0

Recommended Articles

Metrics

Comments