Abstract
Today, Cloud Computing has become the foundation of todays digital world. It allows companies to deploy applications, store huge amounts of data and offer services based on a flexible, on-demand computing model. Unfortunately, today there is still a problem in effectively managing the allocation of Cloud Computing resources. Most companies use traditional ways of scaling their Cloud Computing resources. Some examples of these traditional ways are setting maximum limits for the use of Cloud Computing resources, reviewing historical use of Cloud Computing resources and adjusting to changes in the amount of work being done by the company. There are several problems that could arise from using traditional ways of managing Cloud Computing resources. For instance, you could over-allocate Cloud Computing resources. Also, you could spend too much money on Cloud Computing equipment. Moreover, you could unnecessarily use too much energy. Lastly, you could under-allocate Cloud Computing resources. This paper presents a workload-aware scaling model for managing Cloud Computing resources so that you do not waste money on Cloud Computing resources. The model helps determine whether to allocate, keep, scale up/down, or de-allocate Cloud Computing resources based on how much work is being done, how the use of Cloud Computing resources changes over time, what performance levels the application needs to achieve and how often people want to access the application. It is important to remember that while we want to save money on Cloud Computing resources, we also need to meet service-level objectives, availability and responsiveness. We also propose two ways to maximize the use of Cloud Computing resources and avoid bad scaling decisions. One way is predictive analysis of workload and the other way is adaptive decision making. We create a conceptual evaluation method using four performance metrics. The metrics include; resource allocation efficiency, infrastructure costs, response times, and frequency of scaling events. Instead of saying that we have proven empirically that our model performs better than existing models, we present a research-based model that can be evaluated using either cloud simulation or a controlled deployment experiment. We believe that our results will help develop more cost-effective, responsive, and environmentally-responsible ways of managing Cloud Computing resources.Nonlinear Partial Differential Equations (PDEs) arise from numerous natural and engineering sciences, such as fluid mechanics, quantum physics and heat transfer; their innate complexities include nonlinearity behaviour together with sensitivity to initial and boundary conditions. This problem is formulated in terms of optimizing generalised versions of deep neural networks (DNN) — such universal function approximates lead to a class of methods called physics-informed neural networks (PINNs). PINNs can take advantage of automatic differentiation to calculate the derivatives required for direct calculation of PDE residuals, making them not in need of a specific discretization scheme. Although PINNs have many potential features, there are a few capabilities they lack. However, problems such as training instability, spectral bias, slow convergence and sensitivity to hyper parameters are under-studied. Alternatively, Even when in the same problem type that is required to minimize multiple loss components; PDE residuals and boundary conditions, arise many issues to serial tough problems. Recent advances, including adaptive loss weighting, domain decomposition strategies, and hybrid modelling using data-driven and physics-based approaches are beginning to address the limitations of current practice.
Keywords
Cloud Computing, Workload-Aware Scaling, Resource Allocation, Auto-Scaling, Cloud Optimization, Predictive Analytics, Resource Utilization, Elastic Infrastructure, Cost Efficiency, Performance Management
References
[1] Armbrust, M. et al. (2010) ‘A view of cloud computing’, Communications of the ACM, 53(4), pp. 50–58.
[2] Buyya, R., Yeo, C.S., Venugopal, S., Broberg, J. and Brandic, I. (2009) ‘Cloud computing and emerging IT platforms: Vision, hype, and reality for delivering computing as the 5th utility’, Future Generation Computer Systems, 25(6), pp. 599–616.
[3] Calheiros, R.N., Ranjan, R., Beloglazov, A., De Rose, C.A.F. and Buyya, R. (2011) ‘CloudSim: A toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms’, Software: Practice and Experience, 41(1), pp. 23–50.
[4] Chen, T., Bahsoon, R. and Yao, X. (2018) ‘A survey and taxonomy of self-aware and self-adaptive cloud autoscaling systems’, ACM Computing Surveys, 51(3), Article 61.
[5] Chiang, R.C. and Huang, H.H. (2011) ‘TRACON: Interactively optimizing virtual machine placement in clouds’, in Proceedings of the 2011 IEEE International Conference on Cloud Computing. Washington, DC: IEEE, pp. 81–88.
[6] Dean, J. and Barroso, L.A. (2013) ‘The tail at scale’, Communications of the ACM, 56(2), pp. 74–80.
[7] De Assunção, M.D., da Silva Veith, A. and Buyya, R. (2018) ‘Distributed data processing and caching infrastructure for cloud-based big data analytics’, Future Generation Computer Systems, 86, pp. 1023–1034.
[8] Fan, J., Wang, X., Wu, L. and Liu, J. (2015) ‘An adaptive resource allocation approach for cloud computing based on workload prediction’, International Journal of Distributed Sensor Networks, 11(9).
[9] Galante, G. and de Bona, L.C.E. (2012) ‘A survey on cloud computing elasticity’, in Proceedings of the 2012 IEEE Fifth International Conference on Utility and Cloud Computing. Washington, DC: IEEE, pp. 263–270.
[10] García-Valls, M., Cucinotta, T. and Lu, C. (2014) ‘Challenges in real-time virtualization and predictable cloud computing’, Journal of Systems Architecture, 60(9), pp. 726–740.
[11] Gmach, D., Rolia, J., Cherkasova, L. and Kemper, A. (2007) ‘Workload analysis and demand prediction of enterprise data center applications’, in Proceedings of the 2007 IEEE International Symposium on Workload Characterization. Washington, DC: IEEE, pp. 171–180.
[12] Guitart, J., Carrera, D., Beltran, V., Torres, J. and Ayguadé, E. (2007) ‘A dynamic CPU provisioning model for virtualized service providers’, in Proceedings of the 2007 International Conference on Parallel Processing. Washington, DC: IEEE, pp. 26–33.
[13] Herbst, N.R., Kounev, S. and Reussner, R. (2013) ‘Elasticity in cloud computing: What it is, and what it is not’, in Proceedings of the 10th International Conference on Autonomic Computing. San Jose, CA: USENIX Association, pp. 23–27.
[14] 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.
[15] Jiang, Y. and Perng, C.S. (2010) ‘Cloud resource allocation and management for performance optimization’, Proceedings of the IEEE International Conference on Cloud Computing, pp. 1–8.
[16] Lorido-Botran, T., Miguel-Alonso, J. and Lozano, J.A. (2014) ‘A review of auto-scaling techniques for elastic applications in cloud environments’, Journal of Grid Computing, 12, pp. 559–592.
[17] Mao, M. and Humphrey, M. (2012) ‘A performance study on the VM startup time in the cloud’, in Proceedings of the 2012 IEEE International Conference on Cloud Computing. Washington, DC: IEEE, pp. 423–430.
[18] Masdari, M. and Khoshnevis, A. (2020) ‘A survey and classification of the workload forecasting methods in cloud computing’, Cluster Computing, 23, pp. 1507–1526.
[19] Mauch, V., Kunze, M. and Hillenbrand, M. (2013) ‘High performance cloud computing’, Future Generation Computer Systems, 29(6), pp. 1408–1416.
[20] Qu, C., Calheiros, R.N. and Buyya, R. (2018) ‘Auto-scaling web applications in clouds: A taxonomy and survey’, ACM Computing Surveys, 51(4), Article 73.
[21] Ranjan, R. (2015) ‘Streaming big data processing in datacenter clouds’, IEEE Cloud Computing, 1(1), pp. 78–83.
[22] Rao, J., Bu, X., Xu, C.Z., Wang, L. and Yin, G. (2009) ‘VCONF: A reinforcement learning approach to virtual machines auto-configuration’, in Proceedings of the 6th International Conference on Autonomic Computing and Communications. New York: ACM, pp. 137–146.
[23] Reiss, C., Tumanov, A., Ganger, G.R., Katz, R.H. and Kozuch, M.A. (2012) ‘Heterogeneity and dynamicity of clouds at scale: Google trace analysis’, in Proceedings of the Third ACM Symposium on Cloud Computing. New York: ACM, Article 7, pp. 1–13.
[24] Singh, S. and Chana, I. (2016) ‘A survey on resource scheduling in cloud computing: Issues and challenges’, Journal of Grid Computing, 14, pp. 217–264.
[25] Urgaonkar, B., Pacifici, G., Shenoy, P., Spreitzer, M. and Tantawi, A. (2008) ‘Analytic modeling of multitier Internet applications’, ACM Transactions on the Web, 2(1), Article 1.
[26] Verma, A., Pedrosa, L., Korupolu, M., Oppenheimer, D., Tune, E. and Wilkes, J. (2015) ‘Large-scale cluster management at Google with Borg’, in Proceedings of the 10th European Conference on Computer Systems. New York: ACM, Article 18, pp. 1–17.
[27] Wang, W., Niu, D., Li, B. and Liang, B. (2012) ‘Dynamic cloud resource allocation: Models and algorithms’, in Proceedings of the 2012 IEEE International Conference on Cloud Computing. Washington, DC: IEEE, pp. 1–8.
[28] Xu, J. and Fortes, J.A.B. (2011) ‘Multi-objective virtual machine placement in virtualized data center environments’, in Proceedings of the 2011 IEEE/ACM International Conference on Green Computing and Communications. Washington, DC: IEEE, pp. 179–188.
[29] Zhu, X., Young, D., Watson, B.J., Wang, Z., Rolia, J., Singhal, S., McKee, B., Hyser, C., Gmach, D., Gardner, R., Christian, T. and Cherkasova, L. (2008) ‘1000 islands: An integrated approach to resource management for virtualized data centers’, Cluster Computing, 11, pp. 273–286.
[30] Jeong, B. and Jeong, Y.S. (2025) ‘Autoscaling techniques in cloud-native computing: A comprehensive survey’, Computer Science Review, 58, Article 100791.