Abstract
As the computational needs of edge infrastructures increased, efficient resource management becomes a necessity. An accurate prediction of future resource usage provides insight into dynamic task offloading, proactive auto-scaling, virtual machine migration, and workload balancing. In this paper we propose the use of multi-output one-dimensional convolutional neural networks as resource usage predictors. Convolutional neural networks can manipulate resource usage observations as time series data with the advantage of an adaptive window size selection. In addition, we propose an innovative hybrid hyper-parameter optimization method that combines particle swarm optimization and Bayesian optimization in order to conclude to a close to optimal convolutional neural network architecture. To validate the efficiency of our approach, we conducted experiments with an edge computing infrastructure. The evaluation results show that the proposed regression model achieves higher accuracy as compared to other machine learning meta-predictors and state of the are resource usage models.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Rahmanian, A.A., Ghobaei-Arani, M., Tofighy, S.: A learning automata-based ensemble resource usage prediction algorithm for cloud computing environment. Futur. Gener. Comput. Syst. 79, 54–71 (2018). https://doi.org/10.1016/j.future.2017.09.049
Amiri, M., Mohammad-Khanli, L.: Survey on prediction models of applications for resources provisioning in cloud. J. Netw. Comput. Appl. 82, 93–113 (2017). https://doi.org/10.1016/j.jnca.2017.01.016
Liu, C., Liu, C., Shang, Y., Chen, S., Cheng, B., Chen, J.: An adaptive prediction approach based on workload pattern discrimination in the cloud. J. Netw. Comput. Appl. 80, 35–44 (2017). https://doi.org/10.1016/j.jnca.2016.12.017. ISSN 1084–8045
Calzarossa, M.C., Massari, L., Tessera, D.: Workload characterization: a survey revisited. ACM Comput. Surv. 48(3), 1–43 (2016). https://doi.org/10.1145/2856127. Article 48
Kousiouris, G., Cucinotta, T., Varvarigou, T.: The effects of scheduling, workload type and consolidation scenarios on virtual machine performance and their prediction through optimized artificial neural networks. J. Syst. Softw. 84(8), 1270–1291 (2011). https://doi.org/10.1016/j.jss.2011.04.013. ISSN 0164-1212
Sadeka, I., Jacky, K., Kevin, L., Anna, L.: Empirical prediction models for adaptive resource provisioning in the cloud. Future Gener. Comput. Syst. 28(1), 155–162 (2012). https://doi.org/10.1016/j.future.2011.05.027. ISSN 0167-739X
Litke, A., Tserpes, K., Varvarigou, T.: Computational workload prediction for grid oriented industrial applications: the case of 3D-image rendering. In: CCGrid 2005. IEEE International Symposium on Cluster Computing and the Grid, Cardiff, Wales, UK, vol. 2, pp. 962–969 (2005). https://doi.org/10.1109/CCGRID.2005.1558665
Calheiros, R.N., Masoumi, E., Ranjan, R., Buyya, R.: Workload prediction using Arima model and its impact on cloud applications’ QoS. IEEE Trans. Cloud Comput. 3(4), 449–458 (2015). https://doi.org/10.1109/TCC.2014.2350475
Zhang, Q., Yang, L.T., Yan, Z., Chen, Z., Li, P.: An efficient deep learning model to predict cloud workload for industry informatics. IEEE Trans. Industr. Inf. 14(7), 3170–3178 (2018). https://doi.org/10.1109/TII.2018.2808910
Tan, J., Dube, P., Meng, X., Zhang, L.: Exploiting resource usage patterns for better utilization prediction. In: 2011 31st International Conference on Distributed Computing Systems Workshops, Minneapolis, MN, pp. 14–19 (2011). https://doi.org/10.1109/ICDCSW.2011.53
Cao, J., Zhang, Q., Shi, W.: Challenges and opportunities in edge computing. In: Edge Computing: A Primer. SCS, pp. 59–70. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-02083-5_5
Zhang, K., Zhu, Y., Leng, S., He, Y., Maharjan, S., Zhang, Y.: Deep learning empowered task offloading for mobile edge computing in urban informatics. IEEE Internet Things J. 6(5), 7635–7647 (2019). https://doi.org/10.1109/JIOT.2019.2903191
Zeng, D., Gu, L., Pan, S., Cai, J., Guo, S.: Resource management at the network edge: a deep reinforcement learning approach. IEEE Netw. 33(3), 26–33 (2019). https://doi.org/10.1109/MNET.2019.1800386
Yuan, Q., Li, J., Zhou, H., Lin, T., Luo, G., Shen, X.: A joint service migration and mobility optimization approach for vehicular edge computing. IEEE Trans. Veh. Technol. 69(8), 9041–9052 (2020). https://doi.org/10.1109/TVT.2020.2999617
Borkowski, M., Schulte, S., Hochreiner, C.: Predicting cloud resource utilization. In: Proceedings of the 9th International Conference on Utility and Cloud Computing, New York, NY, USA, pp. 37–42, December 2016. https://doi.org/10.1145/2996890.2996907
Thonglek, K., Ichikawa, K., Takahashi, K., Iida, H., Nakasan, C.: Improving resource utilization in data centers using an LSTM-based prediction model. In: 2019 IEEE International Conference on Cluster Computing (CLUSTER), pp. 1–8, September 2019. https://doi.org/10.1109/CLUSTER.2019.8891022
Baig, S.-R., Iqbal, W., Berral, J.L., Carrera, D.: Adaptive sliding windows for improved estimation of data center resource utilization. Futur. Gener. Comput. Syst. 104, 212–224 (2020). https://doi.org/10.1016/j.future.2019.10.026
Baig, S., Iqbal, W., Berral, J.L., Erradi, A., Carrera, D.: Adaptive prediction models for data center resources utilization estimation. IEEE Trans. Netw. Serv. Manage. 16(4), 1681–1693 (2019). https://doi.org/10.1109/TNSM.2019.2932840
Kaur, G., Bala, A., Chana, I.: An intelligent regressive ensemble approach for predicting resource usage in cloud computing. J. Parallel Distrib. Comput. 123, 1–12 (2019). https://doi.org/10.1016/j.jpdc.2018.08.008
Yaqub, M., et al.: State-of-the-art CNN optimizer for brain tumor segmentation in magnetic resonance images. Brain Sci. 10(7) (2020). https://doi.org/10.3390/brainsci10070427
GitHub, T. Pagoulatou. https://github.com/titapag/HBPSHPO.git
He, Y., Ma, W.J., Zhang, J.P.: The parameters selection of PSO algorithm influencing on performance of fault diagnosis. In: MATEC Web Conference, vol. 63, p. 02019 (2016). https://doi.org/10.1051/matecconf/20166302019
Brochu, E., Cora, V.M., de Freitas, N.: A Tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning. arXiv:1012.2599 [cs], December 2010
GitHub, G. Rodola’, giampaolo/psutil. https://github.com/giampaolo/psutil
GitHub, A. K. Mortensen, anderskm/gputil. https://github.com/anderskm/gputil
Violos, J., Psomakelis, E., Tserpes, K., Aisopos, F., Varvarigou, T.: Leveraging user mobility and mobile app services behavior for optimal edge resource utilization. In: Proceedings of the International Conference on Omni-Layer Intelligent Systems, Crete, Greece, pp. 7–12, May 2019. https://doi.org/10.1145/3312614.3312620
Chen, T., Guestrin, C.: XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 785–794, August 2016. https://doi.org/10.1145/2939672.2939785
Feurer, M., Klein, A., Eggensperger, K., Springenberg, J., Blum, M., Hutter, F.: Efficient and robust automated machine learning. In: Cortes, C., Lawrence, N.D., Lee, D.D., Sugiyama, M., Garnett, R. (eds.) Advances in Neural Information Processing Systems, vol. 28, pp. 2962–2970. Curran Associates Inc. (2015)
Violos, J., Psomakelis, E., Danopoulos, D., Tsanakas, S., Varvarigou, T.: Using LSTM neural networks as resource utilization predictors: the case of training deep learning models on the edge. Presented at the 17th International Conference on the Economics of Grids, Clouds, Systems and Services (GECON), Zenodo. https://doi.org/10.5281/zenodo.4274441
Schmitt, M., Wanka, R.: Particle swarm optimization almost surely finds local optima. Theoret. Comput. Sci. 561, 57–72 (2015). https://doi.org/10.1016/j.tcs.2014.05.017
Berkenkamp, F., Schoellig, A.P., Krause, A.: No-regret Bayesian optimization with unknown hyperparameters. J. Mach. Learn. Res. 20(50), 1–24 (2019)
Acknowledgments
This work is part of the ACCORDION project that has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 871793.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Violos, J., Pagoulatou, T., Tsanakas, S., Tserpes, K., Varvarigou, T. (2021). Predicting Resource Usage in Edge Computing Infrastructures with CNN and a Hybrid Bayesian Particle Swarm Hyper-parameter Optimization Model. In: Arai, K. (eds) Intelligent Computing. Lecture Notes in Networks and Systems, vol 284. Springer, Cham. https://doi.org/10.1007/978-3-030-80126-7_40
Download citation
DOI: https://doi.org/10.1007/978-3-030-80126-7_40
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-80125-0
Online ISBN: 978-3-030-80126-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)