Assessing the Performance and Cost-Efficiency of Serverless Computing for Deploying and Scaling AI and ML Workloads in the Cloud

Authors

  • Nikhil Singla, Rajkumar Balasubramanian, Siddhant Benadikar, Rishabh Rajesh Shanbhag, Ugandhar Dasi

Abstract

This study investigates the efficacy of serverless computing for deploying and scaling artificial intelligence (AI) and machine learning (ML) workloads in cloud environments. We employ a comprehensive methodology to assess performance and cost-efficiency, conducting experiments using popular AI/ML frameworks on leading serverless platforms. Key performance indicators such as latency, throughput, and scalability are measured, alongside an in-depth cost analysis considering resource utilization, operational costs, and total cost of ownership. Our findings reveal that serverless computing offers significant advantages in scalability and cost-efficiency for certain AI/ML workloads, particularly those with intermittent computational needs. However, limitations such as cold start latencies and resource constraints are identified. This research contributes valuable insights for practitioners and researchers, informing decision-making processes for organizations considering serverless computing for AI/ML initiatives.

Downloads

Download data is not yet available.

References

Castro, P., et al. (2019). The rise of serverless computing. Communications of the ACM, 62(12), 44-54.

Amazon Web Services. (2014). AWS Lambda: Run code without thinking about servers. Retrieved from https://aws.amazon.com/lambda/

McGrath, G., & Brenner, P. R. (2017). Serverless computing: Design, implementation, and performance. In 2017 IEEE 37th International Conference on Distributed Computing Systems Workshops (ICDCSW) (pp. 405-410). IEEE.

Baldini, I., et al. (2017). Serverless computing: Current trends and open problems. In Research Advances in Cloud Computing (pp. 1-20). Springer, Singapore.

Lloyd, W., et al. (2018). Serverless computing: An investigation of factors influencing microservice performance. In 2018 IEEE International Conference on Cloud Engineering (IC2E) (pp. 159-169). IEEE.

Adzic, G., & Chatley, R. (2017). Serverless computing: economic and architectural impact. In Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering (pp. 884-889).

Hellerstein, J. M., et al. (2018). Serverless computing: One step forward, two steps back. arXiv preprint arXiv:1812.03651.

Jonas, E., et al. (2019). Cloud programming simplified: A Berkeley view on serverless computing. arXiv preprint arXiv:1902.03383.

LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444.

Hazelwood, K., et al. (2018). Applied machine learning at facebook: A datacenter infrastructure perspective. In 2018 IEEE International Symposium on High Performance Computer Architecture (HPCA) (pp. 620-629). IEEE.

Bhattacharjee, B., et al. (2017). IBM deep learning service. IBM Journal of Research and Development, 61(4/5), 10:1-10:11.

Jouppi, N. P., et al. (2017). In-datacenter performance analysis of a tensor processing unit. In Proceedings of the 44th Annual International Symposium on Computer Architecture (pp. 1-12).

Polyzotis, N., et al. (2018). Data lifecycle challenges in production machine learning: a survey. ACM SIGMOD Record, 47(2), 17-28.

García-Valls, M., Cucinotta, T., & Lu, C. (2014). Challenges in real-time virtualization and predictable cloud computing. Journal of Systems Architecture, 60(9), 726-740.

Chard, R., et al. (2020). Serverless supercomputing: High performance function as a service for science. arXiv preprint arXiv:2005.08492.

Ishakian, V., Muthusamy, V., & Slominski, A. (2018). Serving deep learning models in a serverless platform. In 2018 IEEE International Conference on Cloud Engineering (IC2E) (pp. 257-262). IEEE.

Feng, L., Kudva, P., Da Silva, D., & Hu, J. (2018). Exploring serverless computing for neural network training. In 2018 IEEE 11th International Conference on Cloud Computing (CLOUD) (pp. 334-341). IEEE.

Manner, J., Endreß, M., Heckel, T., & Wirtz, G. (2018). Cold start influencing factors in function as a service. In 2018 IEEE/ACM International Conference on Utility and Cloud Computing Companion (UCC Companion) (pp. 181-188). IEEE.

Spillner, J. (2020). Serverless Literature Dataset. Zenodo. http://doi.org/10.5281/zenodo.1175423

Carreira, J., et al. (2019). A case for serverless machine learning. In Workshop on Systems for ML and Open Source Software at NeurIPS.

Kim, Y. K., & Kim, Y. (2018). Serverless computing for machine learning. In 2018 IEEE International Conference on Big Data (Big Data) (pp. 3567-3569). IEEE.

Gujarati, A., et al. (2020). Serving DNNs like clockwork: Performance predictability from the bottom up. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20) (pp. 443-462).

Rausch, T., et al. (2021). Serverless distributed machine learning: Revitalizing ML for edge computing. In Proceedings of the 22nd International Middleware Conference (pp. 137-150).

Elgamal, T. (2018). Costless: Optimizing cost of serverless computing through function fusion and placement. In 2018 IEEE/ACM Symposium on Edge Computing (SEC) (pp. 300-312). IEEE.

Hall, A., & Ramachandran, U. (2019). An execution model for serverless functions at the edge. In Proceedings of the International Conference on Internet of Things Design and Implementation (pp. 225-236).

Wang, L., et al. (2018). Peeking behind the curtains of serverless platforms. In 2018 USENIX Annual Technical Conference (USENIX ATC 18) (pp. 133-146).

Wen, J., et al. (2021). Chronos: A serverless framework for complex ML pipelines. In 2021 IEEE 41st International Conference on Distributed Computing Systems (ICDCS) (pp. 1101-1111). IEEE.

Eismann, S., et al. (2020). Serverless applications: Why, when, and how? IEEE Software, 38(1), 32-39.

Schleier-Smith, J., et al. (2021). What serverless computing is and should become: The next phase of cloud computing. Communications of the ACM, 64(5), 76-84.

Gartner. (2021). Magic Quadrant for Cloud Infrastructure and Platform Services. Retrieved from https://www.gartner.com/en/documents/3994015

He, K., et al. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).

Eivy, A. (2017). Be wary of the economics of "Serverless" Cloud Computing. IEEE Cloud Computing, 4(2), 6-12.

Lenarduzzi, V., et al. (2020). Serverless computing: A survey of opportunities, challenges and applications. Information, 11(11), 519.

Kuhlenkamp, J., et al. (2020). An empirical study on function placement and cold starts in serverless architectures. In 2020 IEEE International Conference on Software Architecture (ICSA) (pp. 84-94). IEEE

Downloads

Published

16.04.2023

How to Cite

Nikhil Singla. (2023). Assessing the Performance and Cost-Efficiency of Serverless Computing for Deploying and Scaling AI and ML Workloads in the Cloud. International Journal of Intelligent Systems and Applications in Engineering, 11(5s), 618–630. Retrieved from https://www.ijisae.org/index.php/IJISAE/article/view/6730