Modelo Adaptativo para Previsão de Recursos de Rede em Provedores de Internet Modernos

• Dyego H. L. Oliveira State University of Ceará
• Francisco M. V. Filho State University of Ceará
• Thelmo P. de Araújo State University of Ceará
• Joaquim Celestino Jr. State University of Ceará
• Rafael L. Gomes State University of Ceará

DOI: https://doi.org/10.5753/wgrs.2020.12462

Abstract

Currently, Internet Service Providers (ISPs) tend to evolve to Modern Internet Service Providers (MISPs), addressing situations such as the elastic demand for network resources that can affect the Quality of Service (QoS). A promising approach to deal with elastic demand is the usage of a network traffic prediction technique. However, such techniques do not reach the necessary correction factor when the behavior of the network traffic is not clear (as in elastic demand situations). Within this context, this paper presents an adaptive network prediction model for MISPs that adjusts seasonality and trend and removes time series error cycles according to the behavior observed in network traffic. The results, using a real bandwidth data set, suggest that the proposed model improves the existing prediction models.

Keywords: Demanda Elástica, Previsão de Recursos, Modelo Adaptativo

References

Adhikari, R. and Agrawal, R. K. (2013). An introductory study on time series modeling and forecasting. CoRR, abs/1302.6613.

Aldhyani, T. H. H. and Joshi, M. R. (2016). Integration of time series models with softclustering to enhance network traffic forecasting. In 2016 Second International Conference on Research in Computational Intelligence and Communication Networks (ICR-CICN), pages 212-214.

Bayati, A., Khoa Nguyen, K., and Cheriet, M. (2018). Multiple-step-ahead traffic prediction in high-speed networks. IEEE Communications Letters, 22(12):2447-2450.

Carrion-I-Silvestre, J. L., Sansó-I-Rosselló, A., and Ortufio, M. A. (2001). Unit root andstationarity tests' wedding. Economics Letters, 70(1):1-8.

Chatfield, C. and Yar, M. (1988). Holt-winters forecasting: Some practical issues. Journalof the Royal Statistical Society: Series D (The Statistician), 37(2):129-140.

De O. Schmidt, R., Sadre, R., Melnikov, N., Schônwilder, J., and Pras, A. (2014). Linking network usage patterns to traffic Gaussianity fit. 2014 IFIP Networking Conference, IFIP Networking 2014.

Dokumentov, A. and Hyndman, R. J. (2015). STR : A Seasonal-Trend Decomposition Procedure Based on Regression. (13):23.

Foukas, X., Patounas, G., Elmokashfi, A., and Marina, M. K. (2017). Network slicing in 5g: Survey and challenges. IEEE Communications Magazine, 55(5):94-100. Communications Magazine,53(3):199-207.

Harstead, E. and Sharpe, R. (2015). Forecasting of access network bandwidth demands for aggregated subscribers using monte carlo methods. IEEE Communications Magazine, 53(3):199–207.

Hou, Z., She, C., Li, Y., Quek, T. Q. S., and Vucetic, B. (2018). Burstiness aware band width reservation for uplink transmission in tactile internet. In 2018 IEEE In-ternational Conference on Communications Workshops (ICC Workshops), pages 1-6.

Katris, C. and Daskalaki, S. (2019). Dynamic bandwidth allocation for video trafficusing farima-based forecasting models. Journal of Network and Systems Management,27(1):39-65.

Kwiatkowski, D., Phillips, P. C., Schmidt, P., and Shin, Y. (1992). Testing the null hy-pothesis of stationarity against the alternative of a unit root: How sure are we thateconomic time series have a unit root? Journal of Econometrics, 54(1):159 — 178.

Maleki, A., Nasseri, S., Aminabad, M. S., and Hadi, M. (2018). Comparison of ARIMA and NNAR Models for Forecasting Water Treatment Plant's Influent Characteristics.KSCE Journal of Civil Engineering, 22(9):3233-3245.

Ruan, L., Mondal, S., and Wong, E. (2018). Machine learning based bandwidth predictionin tactile heterogencous access networks. In [EEE INFOCOM 2018 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS), pages 1-2.

Schmidt, R. D. O., Sadre, R., and Pras, A. (2013). Gaussian traffic revisited. 2013 IFIP Networking Conference, IFIP Networking 2013, pages 1-9.

Su, F, Dong, H., Jia, L., Qin, Y., and Tian, Z. (2016). Long-term forecasting oriented tourban expressway traffic situation. Advances in Mechanical Engineering, 8(1):1-16.

Wang, T., Guo, Z., Chen, H., and Liu, W. (2018). Bwmanager: Mitigating denial of service attacks in software-defined networks through bandwidth prediction. IEEE Transactions on Network and Service Management, 15(4):1235-1248.

Yar, M. and Chatfield, C. (1990). Prediction intervals for the holt-winters forecasting procedure. International Journal of Forecasting, 6(1):127-137.

Yoo, W. and Sim, A. (2015). Network bandwidth utilization forecast model on high bandwidth networks. In 2015 International Conference on Computing, Networkingand Communications (ICNC), pages 494-498.