پیش بینی بار کاری ماشین های مجازی به منظور کاهش مصرف انرژی در مراکز داده ابری با استفاده از ترکیب مدل های یادگیری ژرف
الموضوعات :
زینب خداوردیان خداوردیان
1
,
حسین صدر
2
,
مژده نظری سلیماندارابی
3
,
سید احمد عدالت پناه
4
1 - واحد علوم و تحقیقات تهران
2 - هیات علمی
3 - مرکز تحقیقات قلب و عروق
4 - موسسه آموزش عالی آیندگان، تنکابن
الکلمات المفتاحية: پیشبیني بار کاري, مراکز داده ابري, انتخاب ماشین مجازي, شبکه عصبي پیچشي, واحد برگشتي دروازهدار,
ملخص المقالة :
افزایش تقاضا براي کاربردهاي مبتني بر ابر و استفاده ناکارآمد از منابع، موجب مصرف بيرویه انرژي در مراکز داده ابري شده است. مدیریت پویاي منابع در مراکز داده با هدف کاهش مصرف انرژي، از طریق پیشبیني بار کاري ماشین مجازي امکانپذیر است. پیشبیني بار کاري ماشین مجازي این امکان را ميدهد که ماشین مجازي متناسب با درخواست کاربران در زمان مناسب مهاجرت کند و در مصرف انرژي موثر باشد و منابع را به کارآمدترین روش تخصیص دهد. پیشبیني بار کاري ماشین مجازي ميتواند بر اساس الگوي درخواست کاربران باشد براي این منظور ميتوان ماشینهاي مجازي را بر اساس پیشبیني مصرف منابع )به عنوان مثال میانگین مصرف پردازنده( در کلاسهاي حساس یا غیر حساس به تأخیر دستهبندي کرد و سپس، ماشینهاي مجازي متناسب با در خواست کاربران را به آنها اختصاص داد. در واقع پیشبیني بار کاري و تحلیل پیشبیني به عنوان یک فرآیند اولیه براي مدیریت منابع )مانند کاهش تعداد مهاجرت در ادغام پویاي ماشین مجازي( باشد. از این رو در این مقاله از ترکیب شبکه عصبي پیچشي و واحد برگشتي دروازهدار بهمنظور پیشبیني بار کاري ماشینهاي مجازي مایکروسافت آزور استفادهشده است. مجموعه داده آزور یک مجموعه داده داراي برچسب است و بار کاري ماشینهاي مجازي در این مجموعه داده در دو برچسب حساس یا غیر حساس به تأخیر قرار دارند. در این مجموعه داده اکثر ماشینهاي مجازي داراي برچسب غیر حساس به تأخیر ميباشند؛ بنابراین بنابراین توزیع نمونهها در این مجموعه داده به صورت نامتوازن است از ین رو براي رفع این چالش از افزایش تصادفي نمونههاي کلاس اقلیت استفاده شده است. طبق نتایج حاصل از آزمایشها، روش پیشنهادي داراي دقت 42 / 94 است که نشاندهنده برتري مدل پیشنهادي نسبت به سایر مدلهاي پیشین است.
Cloud computing service models are growing rapidly, and inefficient use of resources in cloud data centers leads to high energy consumption and increased costs. Plans of resource allocation aiming to reduce energy consumption in cloud data centers has been conducted using live migration of Virtual Machines (VMs) and their consolidation into the small number of Physical Machines (PMs). However, the selection of the appropriate VM for migration is an important challenge. To solve this issue, VMs can be classified according to the pattern of user requests into Delay-sensitive (Interactive) or Delay-Insensitive classes, and thereafter suitable VMs can be selected for migration. This is possible by virtual machine workload prediction .In fact, workload predicting and predicting analysis is a pre-migration process of a virtual machine. In this paper, In order to classification of VMs in the Microsoft Azure cloud service, a hybrid model based on Convolution Neural Network (CNN) and Gated Recurrent Unit (GRU) is proposed. Microsoft Azure Dataset is a labeled dataset and the workload of virtual machines in this dataset are in two labeled Delay-sensitive (Interactive) or Delay-Insensitive. But the distribution of samples in this dataset is unbalanced. In fact, many samples are in the Delay-Insensitive class. Therefore, Random Over-Sampling (ROS) method is used in this paper to overcome this challenge. Based on the empirical results, the proposed model obtained an accuracy of 94.42 which clearly demonstrates the superiority of our proposed model compared to other existing models.
منابع و مأخذ
[1] A. Yousafzai et al., "Cloud resource allocation schemes: review, taxonomy, and opportunities," Knowledge and Information Systems, vol. 50, no. 2, pp. 347-381, 2017.
[2] I. Hamzaoui, B. Duthil, V. Courboulay, and H. Medromi, "A Survey on the Current Challenges of Energy-Efficient Cloud Resources Management," SN Computer Science, vol. 1, no. 2, pp. 1-28, 2020.
[3] A. Beloglazov, "Energy-efficient management of virtual machines in data centers for cloud computing," 2013.
[4] S. Singh and I. Chana, "A survey on resource scheduling in cloud computing: Issues and challenges," Journal of grid computing, vol. 14, no. 2, pp. 217-264, 2016.
[5] M. H. Sayadnavard, A. T. Haghighat, and A. M. Rahmani, "A reliable energy-aware approach for dynamic virtual machine consolidation in cloud data centers," The Journal of Supercomputing, vol. 75, no. 4, pp. 2126-2147, 2019.
[6] J. N. Witanto, H. Lim, and M. Atiquzzaman, "Adaptive selection of dynamic VM consolidation algorithm using neural network for cloud resource management," Future generation computer systems, vol. 87, pp. 35-42, 2018.
[7] H. Sadr, M. M. Pedram, and M. Teshnehlab, "Multi-View Deep Network: A Deep Model Based on Learning Features From Heterogeneous Neural Networks for Sentiment Analysis," IEEE Access, vol. 8, pp. 86984-86997, 2020.
[8] Y. S. Patel and R. Misra, "Performance comparison of deep VM workload prediction approaches for cloud," in Progress in Computing, Analytics and Networking: Springer, 2018, pp. 149-160.
[9] H. Khani and H. Khanmirza, "Randomized routing of virtual machines in IaaS data centers," PeerJ Computer Science, vol. 5, p. e211, 2019.
[10] A. Beloglazov, J. Abawajy, and R. Buyya, "Energy-aware resource allocation heuristics for efficient management of data centers for cloud computing," Future generation computer systems, vol. 28, no. 5, pp. 755-768, 2012.
[11] R. Moreno-Vozmediano, R. S. Montero, E. Huedo, and I. M. Llorente, "Efficient resource provisioning for elastic Cloud services based on machine learning techniques," Journal of Cloud Computing, vol. 8, no. 1, p. 5, 2019.
[12] M. Hariharasubramanian, "Improving application infrastructure provisioning using resource usage predictions from cloud metric data analysis," Rutgers University-School of Graduate Studies, 2018.
[13] R. Shaw, E. Howley, and E. Barrett, "An intelligent ensemble learning approach for energy efficient and interference aware dynamic virtual machine consolidation," Simulation Modelling Practice and Theory, vol. 102, p. 101992, 2020.
[14] N. Verma and A. Sharma, "Workload prediction model based on supervised learning for energy efficiency in cloud," in 2017 2nd International Conference on Communication Systems, Computing and IT Applications (CSCITA), 2017, pp. 66-71: IEEE.
[15] W. Zhang, B. Li, D. Zhao, F. Gong, and Q. Lu, "Workload prediction for cloud cluster using a recurrent neural network," in 2016 International Conference on Identification, Information and Knowledge in the Internet of Things (IIKI), 2016, pp. 104-109: IEEE.
[16] Y. Zhu, W. Zhang, Y. Chen, and H. Gao, "A novel approach to workload prediction using attention-based LSTM encoder-decoder network in cloud environment," EURASIP Journal on Wireless Communications and Networking, vol. 2019, no. 1, p. 274, 2019.
[17] S. Ouhame and Y. Hadi, "Multivariate workload prediction using Vector Autoregressive and Stacked LSTM models," in Proceedings of the New Challenges in Data Sciences: Acts of the Second Conference of the Moroccan Classification Society, 2019, pp. 1-7.
[18] Z. Chen, J. Hu, G. Min, A. Y. Zomaya, and T. El-Ghazawi, "Towards accurate prediction for high-dimensional and highly-variable cloud workloads with deep learning," IEEE Transactions on Parallel and Distributed Systems, vol. 31, no. 4, pp. 923-934, 2019.
[19] P. Yazdanian and S. Sharifian, "Cloud Workload Prediction Using ConvNet And Stacked LSTM," in 2018 4th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS), 2018, pp. 83-87: IEEE.
[20] P. Yazdanian and S. Sharifian, "E2LG: a multiscale ensemble of LSTM/GAN deep learning architecture for multistep-ahead cloud workload prediction," The Journal of Supercomputing, pp. 1-31, 2021.
[21] J. Bi, S. Li, H. Yuan, and M. Zhou, "Integrated deep learning method for workload and resource prediction in cloud systems," Neurocomputing, vol. 424, pp. 35-48, 2021.
[22] S. Shishira and A. Kandasamy, "A Novel Feature Extraction Model for Large-Scale Workload Prediction in Cloud Environment," SN Computer Science, vol. 2, no. 5, pp. 1-7, 2021.
[23] B. Santoso, H. Wijayanto, K. Notodiputro, and B. Sartono, "Synthetic over sampling methods for handling class imbalanced problems: a review," in IOP conference series: earth and environmental science, 2017, vol. 58, no. 1, p. 012031.
[24] E. Jackson and R. Agrawal, "Performance Evaluation of Different Feature Encoding Schemes on Cybersecurity Logs," in 2019 SoutheastCon, 2019, pp. 1-9: IEEE.
[25] H. Sadr, M. M. Pedram, and M. Teshnehlab, "A robust sentiment analysis method based on sequential combination of convolutional and recursive neural networks," Neural Processing Letters, vol. 50, no. 3, pp. 2745-2761, 2019.
[26] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, "Empirical evaluation of gated recurrent neural networks on sequence modeling," arXiv preprint arXiv:1412.3555, 2014.
[27] Y. Wang, M. Liu, Z. Bao, and S. Zhang, "Short-term load forecasting with multi-source data using gated recurrent unit neural networks," Energies, vol. 11, no. 5, p. 1138, 2018.
[28] E. Cortez, A. Bonde, A. Muzio, M. Russinovich, M. Fontoura, and R. Bianchini, "Resource central: Understanding and predicting workloads for improved resource management in large cloud platforms," in Proceedings of the 26th Symposium on Operating Systems Principles, 2017, pp. 153-167.
[29] E. Patel, A. Mohan, and D. S. Kushwaha, "Neural network based classification of virtual machines in IaaS," in 2018 5th IEEE Uttar Pradesh Section International Conference on Electrical, Electronics and Computer Engineering (UPCON), 2018, pp. 1-8: IEEE.
[30] A. M. Aslam and M. Kalra, "Using Artificial Neural Network for VM Consolidation Approach to Enhance Energy Efficiency in Green Cloud," in Advances in Data and Information Sciences: Springer, 2019, pp. 139-154.
[31] M. A. Wani, F. A. Bhat, S. Afzal, and A. I. Khan, Advances in deep learning. Springer, 2020.
[32] A. Plebe and G. Grasso, "The unbearable shallow understanding of deep learning," Minds and Machines, vol. 29, no. 4, pp. 515-553, 2019.
