Comparative Analysis of AWS Model Deployment Services

© 2024 by IJCTT Journal
Volume-72 Issue-5
Year of Publication : 2024
Authors : Rahul Bagai
DOI :  10.14445/22312803/IJCTT-V72I5P113

How to Cite?

Rahul Bagai, "Comparative Analysis of AWS Model Deployment Services," International Journal of Computer Trends and Technology, vol. 72, no. 5, pp. 102-110, 2024. Crossref,

Amazon Web Services (AWS) offers three important Model Deployment Services for model developers: SageMaker, Lambda, and Elastic Container Service (ECS). These services have critical advantages and disadvantages, influencing model developers’ adoption decisions. This comparative analysis reviews the merits and drawbacks of these services. This analysis found that Lambda AWS service leads in efficiency, autoscaling aspects, and integration during model development. However, ECS was found to be outstanding in terms of flexibility, scalability, and infrastructure control; conversely, ECS is better suited when it comes to managing complex container environments during model development, as well as addressing budget concerns- it is, therefore, the preferred option for model developers whose objective is to achieve complete freedom and framework flexibility with horizontal scaling. ECS is better suited to ensuring performance requirements align with project goals and constraints. The AWS service selection process considered factors that include but are not limited to load balance and cost-effectiveness. ECS is a better choice when model development begins from the abstract. It offers unique benefits, such as the ability to scale horizontally and vertically, making it the best preferable tool for model deployment.

AWS ECS, AWS Lambda, AWS SageMaker, Cost Analysis, Machine Learning Deployment, Performance Evaluation, scalability.


[1] T. Aapakallio-Autioet al., “Serverless Computing: A Case Study on AWS Lambda,” IEEE Access, vol. 9, pp. 24284- 24294, 2021.
[2] M. Berg, “An Overview of Amazon SageMaker for Machine Learning,” IEEE Transactions on Services Computing, vol. 5, pp. 183-196, 2022.
[3] Thameez Ahmad Bodhanya, “Comparing Cloud Orchestrated Container Platforms: Under the Lenses of Performance, Cost, Ease-of-Use, and Reliability,” Master Thesis, Uppsala University, pp. 1-45, 2022.
[Google Scholar] [Publisher Link]
[4] S. Choi et al., “Integrating Amazon SageMaker with other AWS Services for Machine Learning Deployment,” IEEE Internet Computing, vol. 27, pp. 28-37, 2023.
[5] Alfredo de Oliveira, “Securing Weak Points in Serverless Architectures,” Resources Trend Micro, pp. 1-31, 2022.
[Google Scholar] [Publisher Link]
[6] M. Hwasser, “Optimizing Machine Learning Deployment with AWS Sagemaker,” IEEE Transactions on Cloud Computing, vol. 10, pp. 584-596, 2022.
[7] S. Kulkarni et al., “Efficient Data Processing Using AWS Glue for Machine Learning Tasks,” IEEE Data Engineering Bulletin, vol. 46, pp. 67-76, 2023.
[8] R. Mathew et al., “Performance Evaluation of AWS Lambda for Machine Learning Workloads,” IEEE Transactions on Services Computing, vol. 14, pp. 927-939, 2021.
[9] K. Muthu, “Evaluating the Cost Implications of Amazon SageMaker for Machine Learning Projects,” IEEE Access, vol. 10, pp. 21796-21808, 2022.
[10] P. Nawagamuwa, “Optimizing Costs with AWS Lambda: A Case Study,” IEEE Cloud Computing, vol. 10, pp. 83-91, 2023.
[11] A. Nigenda et al., “Auto-Scaling Strategies in AWS SageMaker for Machine Learning Deployment,” IEEE Access, vol. 11, pp. 19133-19145, 2022.
[12] A. Pelle et al., “Performance Optimization of AWS Lambda for Real-Time Applications, IEEE Transactions on Cloud Computing, vol. 7, pp. 783-796, 2019.
[13] S. Rajendran et al., Scalability and Efficiency of AWS Lambda for Event-Driven Architectures,” IEEE Transactions on Cloud Computing, vol. 11, pp. 301-313, 2023.
[14] M. Radeck, “Containerization Strategies in Amazon ECS for Machine Learning Applications,” IEEE Transactions on Emerging Topics in Computing, vol. 8, pp. 282-294, 2020.
[15] J. Ramirez et al., “Horizontal Scaling with AWS ECS: A Case Study,” IEEE Transactions on Cloud Computing, vol. 6, pp. 998-1010, 2019.
[16] M. Soncin, “Extending the Capabilities of Amazon SageMaker: A Case Study,” IEEE Software, vol. 40, pp. 59-67, 2023.
[17] Martin Sisák, “Cost-optimal AWS Deployment Configuration for Containerized Event-Driven Systems,” IEEE Transactions on Cloud Computing, vol. 9, no. 4, pp. 1065-1077, 2021.
[Google Scholar]
[18] J. Takkunen, “Flexibility of AWS ECS for Docker containerization,” IEEE Access, vol. 9, pp. 14955-14967, 2021.
[19] K. Trawinski, “Managing Machine Learning Pipelines with Amazon SageMaker,” IEEE Transactions on Big Data, vol. 9, pp. 325-337, 2022.
[20] Y. Xu, “Automated Deployment with AWS ECS: A Case Study,” IEEE Cloud Computing, vol. 7, no. 1, pp. 61-70, 2020.
[21] Abhishek Mishra, Machine Learning in the AWS Cloud: Add Intelligence to Applications with Amazon SageMaker and Amazon Rekognition, Wiley, pp. 1-528, 2019.
[Google Scholar] [Publisher Link]
[22] Dheeraj Chahal et al., “Performance and Cost Comparison of Cloud Services for Deep Learning Workload,” Companion of the ACM/SPEC International Conference on Performance Engineering, France, pp. 49-55, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Can Karakus et al., “Amazon SageMaker Model Parallelism: A General and Flexible Framework for Large Model Training,” Arxiv, pp. 1-24, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Valerio Perrone et al., “Amazon SageMaker Automatic Model Tuning: Scalable Gradient-Free Optimization,” Arxiv, pp. 1-10, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[25] Damiano Perri et al., “Implementing a Scalable and Elastic Computing Environment Based on Cloud Containers,” Computational Science and its Applications–ICCSA 2021: 21st International Conference, Cagliari, Italy, vol. 12949, pp. 676-689, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Khandakar Razoan Ahmed, and Motaharul Islam, “A Comparative Analysis of AWS Cloud-Native Application Deployment Model” Proceedings of International Conference on Fourth Industrial Revolution and Beyond 2021, Springer, Singapore, vol. 437, pp. 429-441, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Jin-young Choi, Minkyoung Cho, and Jik-Soo Kim, “Employing Vertical Elasticity for Efficient Big Data Processing in Container-Based Cloud Environments,” Applied Sciences, vol. 11, no. 13, pp. 1-13, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[28] Sanath Mugeraya, and Kailas Devadkar, “Dynamic Task Scheduling and Resource Allocation for Microservices in Cloud,” Journal of Physics: Conference Series, vol. 2325, pp. 1-10, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[29] Parul Dubey et al., Amazon Web Services: the Definitive Guide for Beginners and Advanced Users, Bentham Science Publishers, pp. 1-207, 2023.
[Google Scholar] [Publisher Link]
[30] Manya K. Ravindranathan, D. Sendil Vadivu, and Narendran Rajagopalan, “Cloud-Driven Machine Learning with AWS: A Comprehensive Review of Services,” 2024 International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics, Bangalore, India, pp. 1-8, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[31] Myungjun Son et al., “Splice: An Automated Framework for Cost-and Performance-Aware Blending of Cloud Services,” 2022 22nd IEEE International Symposium on Cluster, Cloud and Internet Computing, Taormina, Italy, pp. 119-128, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[32] Archana Modawal, How to Deploy Machine Learning Models Using Amazon SageMaker, Softwebsolutions, 2023. [Online]. Available:
[33] Adam Novotný, AWS Lambda Pricing: Cost Optimization Approaches Guide, Stormit. [Online]. Available:
[34] What is Amazon Elastic Container Service?, Amazon Web Services, 2024. [Online]. Available:
[35] Jagreet Kaur Gill, Amazon SageMaker: End-to-End Managed Machine Learning Platform, Xenonstack, 2023. [Online]. Available:
[36] Ben Brostoff, AWS Explained to Shareholders from a Developer, Medium, 2019. [Online]. Available: