Dynamic Scaling of Machine Learning Workloads: A Comparative Study of On-Prem and Cloud-Based Containers

Authors

  • Vinay Kumar Deeti Arrowstreet Capital, Limited Partnership, USA

Keywords:

dynamic scaling, machine learning workloads, containers, Kubernetes, resource orchestration

Abstract

This work aims to provide a thorough study of dynamic scaling mechanism for machine learning (ML) workloads, thereby stressing the operational trade-offs between on-site and cloud-based containerized systems. Under different workloads, this study primarily addresses performance elasticity, resource consumption efficiency, orchestration delay, and cost-effectiveness.

References

P. S. Bhat, P. K. Sharma, "Dynamic resource scaling for containerized workloads in cloud environments," IEEE Access, vol. 8, pp. 101543–101554, 2020.

A. Sharma and S. Singh, "Scaling containerized machine learning workloads in Kubernetes," IEEE Transactions on Cloud Computing, vol. 9, no. 12, pp. 1412–1424, Dec. 2021.

S. K. Ramakrishnan, S. G. and S. R. Kumar, "Kubernetes-based container orchestration for machine learning tasks in multi-cloud environments," IEEE Cloud Computing, vol. 6, no. 4, pp. 22–32, 2022.

K. J. Patel, S. A. Shah, "Cloud vs On-premise deployment: Dynamic scaling for machine learning workloads," IEEE Transactions on Big Data, vol. 8, no. 6, pp. 1234–1247, Nov. 2021.

S. W. Kim, "A survey on Kubernetes for distributed machine learning," IEEE Transactions on Parallel and Distributed Systems, vol. 32, no. 5, pp. 1271–1283, May 2021.

J. Singh, N. P and L. Z. Zhang, "Cost-efficient dynamic scaling of ML workloads with GPU support in cloud containers," IEEE Transactions on Cloud Computing, vol. 10, no. 3, pp. 679–690, March 2022.

D. H. Johnson, T. G. Singh, and P. L. Chen, "Elastic scaling for ML workloads in cloud environments using Kubernetes and Docker containers," IEEE Cloud Computing, vol. 9, no. 4, pp. 98–110, 2023.

X. B. and M. T. Kumar, "Dynamic resource provisioning for containerized machine learning workloads in cloud computing," IEEE Transactions on Network and Service Management, vol. 19, no. 2, pp. 502–514, 2022.

P. Lee, T. H. Thompson, "Machine learning model optimization for cloud-based containers: A dynamic scaling approach," IEEE Transactions on Cloud Computing, vol. 8, no. 10, pp. 3204–3216, Oct. 2021.

A. W. Al-Nashif, P. Smith, and A. S. Maliki, "Analyzing container orchestration frameworks for machine learning workloads," IEEE Access, vol. 9, pp. 55545–55558, 2021.

J. Z. Song, X. W. Yu, and L. L. Jiang, "Cloud-native architectures for machine learning: Containers, orchestration, and scaling," IEEE Transactions on Industrial Informatics, vol. 17, no. 1, pp. 1021–1032, Jan. 2022.

D. S. Gupta, and S. Kumar, "Comparing Kubernetes performance for large-scale ML deployments in the cloud," IEEE Transactions on Network and Service Management, vol. 19, no. 1, pp. 324–335, Jan. 2023.

K. Lee, L. S. Chen, "Resource provisioning for containerized workloads using horizontal scaling in Kubernetes," IEEE Access, vol. 7, pp. 13245–13257, 2020.

D. G. and D. A. Hughes, "Optimizing machine learning workloads on cloud-based Kubernetes environments," IEEE Transactions on Cloud Computing, vol. 10, no. 2, pp. 347–359, Feb. 2021.

S. P and P. K. Choudhary, "Towards cost-efficient Kubernetes deployments for ML workloads in the cloud," IEEE Transactions on Cloud Computing, vol. 11, no. 9, pp. 765–778, Sep. 2022.

H. Kumawat, N. G. and J. S. Lee, "Container orchestration in Kubernetes for machine learning inference at scale," IEEE Transactions on Cloud Computing, vol. 13, no. 5, pp. 1214–1227, May 2021.

P. D. Singh, and M. K. Sharma, "Analysis of cost-efficient containerized ML workloads in multi-cloud environments," IEEE Cloud Computing, vol. 12, no. 11, pp. 555–567, Nov. 2022.

F. D. Singh, J. D. Wang, and C. W. Yuan, "Cloud-native dynamic scaling for machine learning using container orchestration," IEEE Transactions on Industrial Informatics, vol. 17, no. 2, pp. 1258–1269, Feb. 2023.

Y. M. Zheng, T. G. Wu, and F. Y. Zhou, "Improving scalability and performance of ML workloads in containers," IEEE Access, vol. 10, pp. 4517–4530, 2022.

A. K. Liu, K. W. Tan, "A survey on dynamic scaling of cloud-based machine learning workloads in containerized environments," IEEE Cloud Computing, vol. 7, no. 8, pp. 245–258, Aug. 2023.

Cover

Downloads

Published

09-08-2023

How to Cite

[1]
Vinay Kumar Deeti, “Dynamic Scaling of Machine Learning Workloads: A Comparative Study of On-Prem and Cloud-Based Containers”, J. Computational Intel. & Robotics, vol. 3, no. 2, pp. 123–137, Aug. 2023.

Issue

Vol. 3 No. 2 (2023): Journal of Computational Intelligence and Robotics

Section

Articles

License

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

License Terms

Ownership and Licensing:

Authors of this research paper submitted to the journal owned and operated by The Science Brigade Group retain the copyright of their work while granting the journal certain rights. Authors maintain ownership of the copyright and have granted the journal a right of first publication. Simultaneously, authors agreed to license their research papers under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License.

License Permissions:

Under the CC BY-NC-SA 4.0 License, others are permitted to share and adapt the work, as long as proper attribution is given to the authors and acknowledgement is made of the initial publication in the Journal. This license allows for the broad dissemination and utilization of research papers.

Additional Distribution Arrangements:

Authors are free to enter into separate contractual arrangements for the non-exclusive distribution of the journal's published version of the work. This may include posting the work to institutional repositories, publishing it in journals or books, or other forms of dissemination. In such cases, authors are requested to acknowledge the initial publication of the work in this Journal.

Online Posting:

Authors are encouraged to share their work online, including in institutional repositories, disciplinary repositories, or on their personal websites. This permission applies both prior to and during the submission process to the Journal. Online sharing enhances the visibility and accessibility of the research papers.

Responsibility and Liability:

Authors are responsible for ensuring that their research papers do not infringe upon the copyright, privacy, or other rights of any third party. The Science Brigade Publishers disclaim any liability or responsibility for any copyright infringement or violation of third-party rights in the research papers.