Fault-Tolerant Resource Optimization using Bi-LSTM with Attention in Cloud Computing

doi:10.23940/ijpe.25.09.p4.506520

Abstract

Abstract: This paper introduces a novel approach for forecasting CPU utilization in cloud computing environments using a Bi-LSTM model enhanced with an attention mechanism. By addressing dataset irregularities with an interpolation-based data preprocessing technique, the method ensures accurate representation and is validated by the Kolmogorov-Smirnov two-sample test. The attention mechanism within the Bi-LSTM model improves prediction accuracy by identifying key dataset features. Evaluated using the Alibaba Cluster Trace dataset, the approach demonstrates superior performance compared to established methods. Predicted utilization values facilitate fault-tolerant VM allocation and resource extension, reducing overutilized host occurrences and enhancing system reliability. This predictive allocation minimizes unnecessary VM migrations, decreases overhead, and ensures optimal host utilization, leading to more reliable services and improved adherence to SLAs.

Key words: cloud computing, fault tolerance, resource allocation, virtual machine

Neetu Narang Mahajan and Parmeet Kaur. Fault-Tolerant Resource Optimization using Bi-LSTM with Attention in Cloud Computing [J]. Int J Performability Eng, 2025, 21(9): 506-520.

Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks

References

[1] Mukwevho M.A., and Celik T., 2018. Toward a smart cloud: A review of fault-tolerance methods in cloud systems. IEEE Transactions on Services Computing,14(2), pp. 589-605.
[2] Sotiriadis S., Bessis N., Petrakis E.G., Amza C., Negru C., and Mocanu M., 2017. Virtual machine cluster mobility in inter-cloud platforms.Future Generation Computer Systems, 74, pp. 179-189.
[3] Aslanpour M.S., Gill S.S., and Toosi A.N., 2020. Performance evaluation metrics for cloud, fog and edge computing: A review, taxonomy, benchmarks and stand ards for future research.Internet of Things, 12, 100273.
[4] Shahid M.A., Islam N., Alam M.M., Mazliham M.S., and Musa S., 2021. Towards resilient method: an exhaustive survey of fault tolerance methods in the cloud computing environment.Computer Science Review, 40, 100398.
[5] Kumari P., and Kaur P., 2021. A survey of fault tolerance in cloud computing. Journal of King Saud University-Computer and Information Sciences,33(10), pp. 1159-1176.
[6] Polze A., Tröger P., and Salfner F., 2011. Timely virtual machine migration for pro-active fault tolerance. In2011 14th IEEE International Symposium on Object/Component/Service-Oriented Real-Time Distributed Computing Workshops, pp. 234-243.
[7] Zhang F., Liu G., Fu X., and Yahyapour R., 2018. A survey on virtual machine migration: challenges, techniques, and open issues. IEEE Communications Surveys & Tutorials,20(2), pp. 1206-1243.
[8] Liu H.,2011. A measurement study of server utilization in public clouds. In2011 IEEE Ninth International Conference on Dependable, Autonomic and Secure Computing, pp. 435-442.
[9] Tuli S., Casale G., and Jennings N.R., 2023. PreGAN+: semi-supervised fault prediction and preemptive migration in dynamic mobile edge environments. IEEE Transactions on Mobile Computing,23(6), pp. 6881-6895.
[10] Tuli S., Casale G., and Jennings N.R., 2022. Pregan: preemptive migration prediction network for proactive fault-tolerant edge computing. InIEEE INFOCOM 2022-IEEE Conference on Computer Communications, pp. 670-679.
[11] Shen H., and Hong X., 2020. Host load prediction with bi-directional long short-term memory in cloud computing.Arxiv Preprint Arxiv:2007.15582.
[12] A. El-Moursy A., Abdelsamea A., Kamran R., and Saad M., 2019. Multi-dimensional regression host utilization algorithm (MDRHU) for host overload detection in cloud computing.Journal of Cloud Computing, 8(1), 8.
[13] Weerakody P.B., Wong K.W., Wang G., and Ela W., 2021. A review of irregular time series data hand ling with gated recurrent neural networks.Neurocomputing, 441, pp. 161-178.
[14] Petrusevich D.A.,2021. Review of missing values procession methods in time series data. In Journal of Physics: Conference Series(Vol. 1889, No. 3, 032009.
[15] Liu J., Wang S., Zhou A., Kumar S.A., Yang F., and Buyya R., 2016. Using proactive fault-tolerance approach to enhance cloud service reliability. IEEE Transactions on Cloud Computing,6(4), pp. 1191-1202.
[16] Huang J., Jin H., Xie X., and Zhang Q., 2007. Using NARX neural network based load prediction to improve scheduling decision in grid environments. InThird International Conference on Natural Computation (ICNC 2007), 5, pp. 718-724.
[17] Roy N., Dubey A., and Gokhale A., 2011. Efficient autoscaling in the cloud using predictive models for workload forecasting. In2011 IEEE 4th International Conference on Cloud Computing, pp. 500-507.
[18] Hu Y., Deng B., and Peng F., 2016. Autoscaling prediction models for cloud resource provisioning. In2016 2nd IEEE International Conference on Computer and Communications (ICCC), pp. 1364-1369.
[19] Wu Y., Yuan Y., Yang G., and Zheng W., 2007. Load prediction using hybrid model for computational grid. In2007 8th IEEE/ACM International Conference on Grid Computing, pp. 235-242.
[20] Chen J., and Wang Y., 2019. A hybrid method for short‐term host utilization prediction in cloud computing.Journal of Electrical and Computer Engineering, 2019(1), 2782349.
[21] Kim I.K., Wang W., Qi Y., and Humphrey M., 2018. Cloudinsight: utilizing a council of experts to predict future cloud application workloads. In2018 IEEE 11th International Conference on Cloud Computing (CLOUD), pp. 41-48.
[22] Yang L., Foster I., and Schopf J.M., 2003. Homeostatic and tendency-based CPU load predictions. InProceedings International Parallel and Distributed Processing Symposium, pp. 9.
[23] Dinda P.A., and O'hallaron D.R., 1999. An evaluation of linear models for host load prediction. InProceedings. the Eighth International Symposium on High Performance Distributed Computing (Cat. No. 99TH8469), pp. 87-96.
[24] Hsieh S.Y., Liu C.S., Buyya R., and Zomaya A.Y., 2020. Utilization-prediction-aware virtual machine consolidation approach for energy-efficient cloud data centers.Journal of Parallel and Distributed Computing, 139, pp. 99-109.
[25] Li L., Dong J., Zuo D., and Wu J., 2019. SLA-aware and energy-efficient VM consolidation in cloud data centers using robust linear regression prediction model.IEEE Access, 7, pp. 9490-9500.
[26] Prevost J.J., Nagothu K., Kelley B., and Jamshidi M., 2011. Prediction of cloud data center networks loads using stochastic and neural models. In2011 6th International Conference on System of Systems Engineering, pp. 276-281.
[27] Duggan M., Mason K., Duggan J., Howley E., and Barrett E., 2017. Predicting host CPU utilization in cloud computing using recurrent neural networks. In2017 12th International Conference for Internet Technology and Secured Transactions (ICITST), pp. 67-72.
[28] Islam S., Keung J., Lee K., and Liu A., 2012. Empirical prediction models for adaptive resource provisioning in the cloud. Future Generation Computer Systems,28(1), pp. 155-162.
[29] Duy T.V.T., Sato Y., and Inoguchi Y., 2011. Improving accuracy of host load predictions on computational grids by artificial neural networks. International Journal of Parallel, Emergent and Distributed Systems,26(4), pp. 275-290.
[30] Peng C., Li Y., Yu Y., Zhou Y., and Du S., 2018. Multi-step-ahead host load prediction with gru based encoder-decoder in cloud computing. In2018 10th International Conference on Knowledge and Smart Technology (KST), pp. 186-191.
[31] Bey K.B., Benhammadi F., Mokhtari A., and Guessoum Z., 2009. CPU load prediction model for distributed computing. In2009 Eighth International Symposium on Parallel and Distributed Computing, pp. 39-45.
[32] Kasabov N.K., and Song Q., 2002. DENFIS: dynamic evolving neural-fuzzy inference system and its application for time-series prediction. IEEE Transactions on Fuzzy Systems,10(2), pp. 144-154.
[33] Mason K., Duggan M., Barrett E., Duggan J., and Howley E., 2018. Predicting host CPU utilization in the cloud using evolutionary neural networks.Future Generation Computer Systems, 86, pp. 162-173.
[34] Lepot M., Aubin J.B., and Clemens F.H., 2017. Interpolation in time series: an introductive overview of existing methods, their performance criteria and uncertainty assessment.Water, 9(10), 796.
[35] Lin P.C., Wu B., and Watada J., 2010. Kolmogorov-smirnov two sample test with continuous fuzzy data. InIntegrated Uncertainty Management and Applications, pp. 175-186.
[36] Wang Y., Huang M., Zhu X., and Zhao L., 2016. Attention-based LSTM for aspect-level sentiment classification. InProceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pp. 606-615.
[37] Srikantaiah S., Kansal A., and Zhao F., 2008. Energy aware consolidation for cloud computing. InUSENIX HotPower'08: Workshop on Power Aware Computing and Systems at OSDI.
[38] Beloglazov A., and Buyya R., 2012. Managing overloaded hosts for dynamic consolidation of virtual machines in cloud data centers under quality of service constraints. IEEE Transactions on Parallel and Distributed Systems,24(7), pp. 1366-1379.
[39] Sun C., Hong S., Song M., and Li H., 2020. A review of deep learning methods for irregularly sampled medical time series data.Arxiv Preprint Arxiv:2010.12493.
[40] Aslam M.,2019. Introducing kolmogorov-smirnov tests under uncertainty: an application to radioactive data. ACS Omega,5(1), pp. 914-917.
[41] Barquera R., Collen E., Di D., Buhler S., Teixeira J., Llamas B., Nunes J.M., and Sanchez‐Mazas A., 2020. Binding affinities of 438 HLA proteins to complete proteomes of seven pand emic viruses and distributions of strongest and weakest HLA peptide binders in populations worldwide. Hla,96(3), pp. 277-298.
[42] Tang M., and Pan S., 2015. A hybrid genetic algorithm for the energy-efficient virtual machine placement problem in data centers. Neural Processing Letters,41(2), pp. 211-221.
[43] Kim J., and Moon N., 2019. BiLSTM model based on multivariate time series data in multiple field for forecasting trading area.Journal of Ambient Intelligence and Humanized Computing, pp. 1-10.
[44] Xu H., Xu S., Wei W., and Guo N., 2023. Fault tolerance and quality of service aware virtual machine scheduling algorithm in cloud data centers. the Journal of Supercomputing,79(3), pp. 2603-2625.
[45] Alibaba, clusterdata, https://github.com/alibaba/clusterdata/, accessed on August 1, 2025.
[46] Shcherbakov M.V., Brebels A., Shcherbakova N.L., Tyukov A.P., Janovsky T.A., and Kamaev V.A.E., 2013. A survey of forecast error measures. World Applied Sciences Journal,24(24), pp. 171-176.
[47] Sathiyamoorthi V., Keerthika P., Suresh P., and Logeswaran K., 2021. Adaptive fault tolerant resource allocation scheme for cloud computing environments. Journal of Organizational and End User Computing (JOEUC),33(5), pp. 135-152.
[48] Yadav R., Zhang W., Li K., Liu C., and Laghari A.A., 2021. Managing overloaded hosts for energy-efficiency in cloud data centers. Cluster Computing,24(3), pp. 2001-2015.
[49] Khan M.S.A., and Santhosh R., 2022. Hybrid optimization algorithm for vm migration in cloud computing.Computers and Electrical Engineering, 102, 108152.
[50] Gupta P., Sahoo P.K., and Veeravalli B., 2021. Dynamic fault tolerant scheduling with response time minimization for multiple failures in cloud.Journal of Parallel and Distributed Computing, 158, pp. 80-93.