CoChat
Sign Up
Cited by 142Cloud Computing and Resource ManagementEnergy Efficiency in ComputingSoftware-Defined Networks and 5G

esDNN: Deep Neural Network Based Multivariate Workload Prediction in Cloud Computing Environments

Open access
Xu, Minxian, Song, Chenghao, Wu, Huaming, Gill, Sukhpal Singh, Ye, Kejiang, Xu, Chengzhong·2022·ACM Transactions on Internet Technology
DOI 10.1145/3524114
Open this paper in CoChat — free accountOpen PDF

Abstract

Cloud computing has been regarded as a successful paradigm for IT industry by providing benefits for both service providers and customers. In spite of the advantages, cloud computing also suffers from distinct challenges, and one of them is the inefficient resource provisioning for dynamic workloads. Accurate workload predictions for cloud computing can support efficient resource provisioning and avoid resource wastage. However, due to the high-dimensional and high-variable features of cloud workloads, it is difficult to predict the workloads effectively and accurately. The current dominant work for cloud workload prediction is based on regression approaches or recurrent neural networks, which fail to capture the long-term variance of workloads. To address the challenges and overcome the limitations of existing works, we proposed an e fficient supervised learning-based D eep N eural Network ( esDNN ) approach for cloud workload prediction. First, we utilize a sliding window to convert the multivariate data into a supervised learning time series that allows deep learning for processing. Then, we apply a revised Gated Recurrent Unit (GRU) to achieve accurate prediction. To show the effectiveness of esDNN, we also conduct comprehensive experiments based on realistic traces derived from Alibaba and Google cloud data centers. The experimental results demonstrate that esDNN can accurately and efficiently predict cloud workloads. Compared with the state-of-the-art baselines, esDNN can reduce the mean square errors significantly, e.g., 15%. rather than the approach using GRU only. We also apply esDNN for machines auto-scaling, which illustrates that esDNN can reduce the number of active hosts efficiently, thus the costs of service providers can be optimized.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1145/3524114

Cites

2 of 2 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • AI at the edge requires massively dispersed computing capacity.

    “AI at the edge needs massively dispersed computing capacity [33].”

    From Edge AI: A survey (2023) · cites this paper as [33]

    The paper’s cloud-workload discussion and Section 5.4 address predicting workloads and adjusting the number of machines in cloud data centers, not capacity requirements for AI at the edge. Searches for edge AI and dispersed computing capacity found no passage in our copy making the attributed claim. (LLM, conf=0.76)

    Experimental — accuracy not yet measured

  • Edge computing applications provide benefits where sensing, pre-processing, and segmentation can be done at end devices or the edge network.

    “These applications provide benefits where sensing, pre-processing and segmentation can be done at the end devices or the edge network [33].”

    From Edge AI: A survey (2023) · cites this paper as [33]

    The Conclusions and Future Work section proposes extending esDNN to edge computing to reduce response time through offloading. Our copy does not address sensing, pre-processing, or segmentation at end devices or the edge network; searches for those activities found no supporting passage. (LLM, conf=0.76)

    Experimental — accuracy not yet measured

  • The mobile MEC sector is more challenging, with hardware, software, and functional architecture requirements only emerging from demonstration projects.

    “This is a more challenging sector with requirements for hardware and software and functional architectures that are only emerging from demonstration projects [33].”

    From Edge AI: A survey (2023) · cites this paper as [33]

    The Conclusions and Future Work section mentions a possible future extension to edge computing, but says nothing about mobile MEC hardware, software, or functional architecture requirements emerging from demonstration projects. Searches for the quoted wording and for mobile MEC demonstration projects found no passage addressing that claim. (LLM, conf=0.75)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 40

  1. [1][1] Minxian Xu and Rajkumar Buyya. Brownout approach for adaptive management of resources and applications in cloud computing systems: A taxonomy and future directions. ACM Computing Surveys, 52(1):8:1–8:27, 2019.
  2. [2][2] Rajkumar Buyya, Chee Shin Yeo, and Srikumar Venugopal. Market-oriented cloud computing: Vision, hype, and reality for delivering it services as computing utilities. In Proceedings of the 10th IEEE International Conference on High Performance Computing and Communications, pages 5–13, 2008.
  3. [3][3] Mohammad Hossein Ghahramani, MengChu Zhou, and Chi Tin Hon. Toward cloud computing qos architecture: Analysis of cloud systems and cloud services. IEEE/CAA Journal of Automatica Sinica, 4(1):6–18, 2017.
  4. [4][4] M. Du, Y. Wang, K. Ye, and C. Xu. Algorithmics of cost-driven computation offloading in the edge-cloud environment. IEEE Transactions on Computers, 69(10):1519–1532, 2020.
  5. [5][5] Yujun Chen, Xian Yang, Qingwei Lin, Hongyu Zhang, Feng Gao, Zhangwei Xu, Yingnong Dang, Dongmei Zhang, Hang Dong, Yong Xu, Hao Li, and Yu Kang. Outage prediction and diagnosis for cloud service systems. In The World Wide Web Conference, WWW '19, pages 2659–2665, New York, NY, USA, 2019. 1, 1, 1. , March 2022. esDNN: Deep Neural Network based Multivariate Workload Prediction Approach in Cloud Environments 1:23
  6. [6][6] S. Wang, X. Li, and R. Ruiz. Performance analysis for heterogeneous cloud servers using queueing theory. IEEE Transactions on Computers, 69(4):563–576, 2020.
  7. [7][7] 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, 31(4):923–934, April 2020.
  8. [8][8] W. Chen, K. Ye, Y. Wang, G. Xu, and C. Xu. How does the workload look like in production cloud? analysis and clustering of workloads on alibaba cluster trace. In 2018 IEEE 24th International Conference on Parallel and Distributed Systems (ICPADS), pages 102–109, 2018.
  9. [9][9] T. Mikolov, S. Kombrink, L. Burget, J. Černocký, and S. Khudanpur. Extensions of recurrent neural network language model. In 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 5528–5531, 2011.
  10. [10][10] Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural Computation, 9(8):1735–1780, 1997.
  11. [11][11] Junyoung Chung, Caglar Gulcehre, Kyunghyun Cho, and Yoshua Bengio. Empirical evaluation of gated recurrent neural networks on sequence modeling. In NIPS 2014 Workshop on Deep Learning,, December 2014.
  12. [12][12] Hervé Abdi and Lynne J. Williams. Principal component analysis. WIREs Computational Statistics, 2(4):433–459, 2010.
  13. [14][14] Jason Brownlee. Supervised and unsupervised machine learning algorithms. Machine Learning Mastery, 16(03), 2016.
  14. [15][15] R. N. Calheiros, E. Masoumi, R. Ranjan, and R. Buyya. Workload prediction using arima model and its impact on cloud applications' qos. IEEE Transactions on Cloud Computing, 3(4):449–458, Oct 2015.
  15. [16][16] Jingqi Yang, Chuanchang Liu, Yanlei Shang, Bo Cheng, Zexiang Mao, Chunhong Liu, Lisha Niu, and Junliang Chen. A cost-aware auto-scaling approach using the workload prediction in service clouds. Information Systems Frontiers, 16(1):7–18, March 2014.
  16. [17][17] Katja Cetinski and Matjaz B. Juric. Ame-wpc: Advanced model for efficient workload prediction in the cloud. Journal of Network and Computer Applications, 55:191 – 201, 2015.
  17. [18][18] Parminder Singh, Pooja Gupta, and Kiran Jyoti. Tasm: Technocrat arima and svr model for workload prediction of web applications in cloud. Cluster Computing, 22(2):619–633, June 2019.
  18. [19][19] Chunhong Liu, Chuanchang Liu, Yanlei Shang, Shiping Chen, Bo Cheng, and Junliang Chen. An adaptive prediction approach based on workload pattern discrimination in the cloud. Journal of Network and Computer Applications, 80:35 – 44, 2017.
  19. [20][20] J. Bi, H. Yuan, and M. Zhou. Temporal prediction of multiapplication consolidated workloads in distributed clouds. IEEE Transactions on Automation Science and Engineering, 16(4):1763–1773, 2019.
  20. [22][22] Q. Zhang, L. T. Yang, Z. Yan, Z. Chen, and P. Li. An efficient deep learning model to predict cloud workload for industry informatics. IEEE Transactions on Industrial Informatics, 14(7):3170–3178, 2018.
  21. [24][24] F. Qiu, B. Zhang, and J. Guo. A deep learning approach for vm workload prediction in the cloud. In 2016 17th IEEE/ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD), pages 319–324, May 2016.
  22. [25][25] Yonghua Zhu, Weilin Zhang, Yihai Chen, and Honghao Gao. A novel approach to workload prediction using attention- based lstm encoder-decoder network in cloud environment. EURASIP Journal on Wireless Communications and Networking, 2019(1):274, 2019.
  23. [27][27] Eli Cortez, Anand Bonde, Alexandre Muzio, Mark Russinovich, Marcus Fontoura, and Ricardo 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, SOSP '17, page 153–167, New York, NY, USA, 2017.
  24. [29][29] Md Ebtidaul Karim, Mirza Mohd Shahriar Maswood, Sunanda Das, and Abdullah G Alharbi. Bhyprec: A novel bi-lstm based hybrid recurrent neural network model to predict the cpu workload of cloud virtual machine. IEEE Access, 9:131476–131495, 2021.
  25. [30][30] Chuang Chen, Ningyun Lu, Bin Jiang, and Cunsong Wang. A risk-averse remaining useful life estimation for predictive maintenance. IEEE/CAA Journal of Automatica Sinica, 8(2):412–422, 2021. 1, 1, 1. , March 2022. 1:24 Xu et al.
  26. [31][31] Ashutosh Kumar Singh, Deepika Saxena, Jitendra Kumar, and Vrinda Gupta. A quantum approach towards the adaptive prediction of cloud workloads. IEEE Transactions on Parallel and Distributed Systems, 2021.
  27. [32][32] In Kee Kim, Wei Wang, Yanjun Qi, and Marty Humphrey. Forecasting cloud application workloads with cloudinsight for predictive resource management. IEEE Transactions on Cloud Computing, 2020.
  28. [34][34] Xuesong Li, Yating Liu, Kunfeng Wang, and Fei-Yue Wang. A recurrent attention and interaction model for pedestrian trajectory prediction. IEEE/CAA Journal of Automatica Sinica, 7(5):1361–1370, 2020.
  29. [36][36] Junfei Qiao, Fei Li, Cuili Yang, Wenjing Li, and Ke Gu. A self-organizing rbf neural network based on distance concentration immune algorithm. IEEE/CAA Journal of Automatica Sinica, 7(1):276–291, 2019.
  30. [37][37] Thomas G Dietterich. Machine learning for sequential data: A review. In Joint IAPR international workshops on statistical techniques in pattern recognition (SPR) and structural and syntactic pattern recognition (SSPR), pages 15–30. Springer, 2002.
  31. [39][39] Dongkuan Xu, Wei Cheng, Bo Zong, Dongjin Song, Jingchao Ni, Wenchao Yu, Yanchi Liu, Haifeng Chen, and Xiang Zhang. Tensorized lstm with adaptive shared memory for learning trends in multivariate time series. In AAAI, pages 1395–1402, 2020.
  32. [40][40] Lars Buitinck, Gilles Louppe, Mathieu Blondel, Fabian Pedregosa, Andreas Mueller, Olivier Grisel, Vlad Niculae, Peter Prettenhofer, Alexandre Gramfort, Jaques Grobler, Robert Layton, Jake VanderPlas, Arnaud Joly, Brian Holt, and Gaël Varoquaux. API design for machine learning software: experiences from the scikit-learn project. In ECML PKDD Workshop: Languages for Data Mining and Machine Learning, pages 108–122, 2013.
  33. [41][41] Rui Fu, Zuo Zhang, and Li Li. Using lstm and gru neural network methods for traffic flow prediction. In 2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC), pages 324–328. IEEE, 2016.
  34. [42][42] Prajit Ramachandran, Barret Zoph, and Quoc V Le. Searching for activation functions. arXiv preprint arXiv:1710.05941, 2017. · arXiv
  35. [43][43] Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker, Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng. Tensorflow: A system for large-scale machine learning. In 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI 16), pages 265–283, Savannah, GA, November 2016.
  36. [45][45] Hervé Bourlard and Yves Kamp. Auto-association by multilayer perceptrons and singular value decomposition. Biological cybernetics, 59(4):291–294, 1988.
  37. [46][46] Andrew Ng et al. Sparse autoencoder. CS294A Lecture notes, 72(2011):1–19, 2011.
  38. [47][47] Xugang Lu, Yu Tsao, Shigeki Matsuda, and Chiori Hori. Speech enhancement based on deep denoising autoencoder. In Interspeech, volume 2013, pages 436–440, 2013.
  39. [48][48] Prajit Ramachandran, Barret Zoph, and Quoc V. Le. Searching for activation functions. CoRR, abs/1710.05941, 2017. · arXiv
  40. [49][49] Mohammad Sadegh Aslanpour, Mostafa Ghobaei-Arani, and Adel Nadjaran Toosi. Auto-scaling web applications in clouds: A cost-aware approach. Journal of Network and Computer Applications, 95:26 – 41, 2017. 1, 1, 1. , March 2022. http://googleresearch.blogspot.com/2011/11/more-google-cluster-data.html http://googleresearch.blogspot.com/2011/11/more-google-cluster-data.html

Cited by

Sorted by citations

Authors
Xu, M., Song, C., Wu, H., Gill, S. S., Ye, K., & Xu, C.
Affiliations
  • Chinese Academy of Sciences (CN)
  • Tianjin University (CN)
  • Queen Mary University of London (GB)
  • University of Macau (MO)
ORCID
Venue
ACM Transactions on Internet Technology
Publisher
Association for Computing Machinery (ACM)
Published
2022-08-17
Volume
22
Issue
3
Pages
1-24
ISSN
  • 1533-5399 (ISSN-L)
  • 1557-6051
DOI
10.1145/3524114
OpenAlex
W4221005058
Cited by
142
References
2
Citations by year
2026: 32 · 2025: 45 · 2024: 37 · 2023: 19 · 2022: 9
FWCI
27.24
Type
journal article
Language
en
Open access
Licence
www.acm.org/publications/policies/copyright_policy#Background (vor)
Locations
  • ACM Transactions on Internet Technology (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • Queen Mary Research Online (Queen Mary University of London) (submittedVersion)
  • arXiv (Cornell University) (public-domain)
Funders
  • National Key R&D Program of China (2021YFB3300200)
  • National Natural Science Foundation of China (62102408 and 62071327)
  • Shenzhen Basic Research Program (JCYJ20200109115418592)
  • Youth Innovation Promotion Association CAS (2019349)
Topics
  • Physical Sciences › Computer Science › Information Systems › Cloud Computing and Resource Management
  • Physical Sciences › Computer Science › Hardware and Architecture › Energy Efficiency in Computing
  • Physical Sciences › Computer Science › Computer Networks and Communications › Software-Defined Networks and 5G
Keywords
cloud workload prediction, multivariate time series, deep neural network, gated recurrent unit, sliding window, resource provisioning, auto-scaling, cloud data centers, resource cost optimization

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review