Federated Learning and API Architectures: Investigating REST API and GraphQL for Decentralized Predictive Analytics
  IJCTT-book-cover
 		          
 
© 2025 by IJCTT Journal
Volume-73 Issue-3
Year of Publication : 2025
Authors : Ramesh Kasarla
DOI :  10.14445/22312803/IJCTT-V73I3P111

How to Cite?

Ramesh Kasarla, "Federated Learning and API Architectures: Investigating REST API and GraphQL for Decentralized Predictive Analytics," International Journal of Computer Trends and Technology, vol. 73, no. 3, pp. 83-91, 2025. Crossref, https://doi.org/10.14445/22312803/IJCTT-V73I3P111

Abstract
Federated learning (FL) is a game-changing methodology for distributed machine learning that allows training models across many edge devices without centralizing sensitive data. Here, this technique is especially useful for privacy sensitive applications like financial, healthcare, or IoT, where the technique helps solve the issue of predictive analytics. However, the communication and coordination of models between distributed nodes remains a critical issue. In this paper, the role of the API architectures in providing the federation learning workflows with a focus on REST and GraphQL are analyzed. Rest APIs are widely popular because of their simplicity and stateless nature, which lends them to lightweight communication in FL environments.
Nevertheless, GraphQL gives clients more flexibility and greater efficiency by allowing them to just ask for the information they require, rather than fetching all data as in a traditional setup, which is a very important feature for decentralized systems. In the FL context, we consider data synchronization, model aggregation, and access controls and analyze the performance, security, and scalability of the two API paradigms. We also discuss how to best design APIs for such federated model training while meeting data protection regulations. This study compares the flexibility of REST and GraphQL to distributed models of artificial intelligence in providing insights on best practices of API design for decentralized federated prediction.

Keywords
Federated Learning, REST API, GraphQL, Decentralized Machine Learning, Predictive Analytics, API Architectures, Model Aggregation.

Reference

[1] G. Anthony Reina et al., “OpenFL: An Open-Source Framework for Federated Learning,” Arxiv, pp. 1-22, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Nicolas Kourtellis, Kleomenis Katevas, and Diego Perinom, “FLaaS: Federated Learning as a Service,” Proceedings of the 1st Workshop on Distributed Machine Learning, Barcelona Spain, pp. 7-13, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Ivan Kholod et al., “Open-Source Federated Learning Frameworks for IoT: A Comparative Review and Analysis,” Sensors, vol. 21, no. 1, pp. 1-22, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Jan Stuecke, Top 7 Open-Source Frameworks for Federated Learning, APHERIS, 2024. [Online]. Available: https://www.apheris.com/resources/blog/top-7-open-source-frameworks-for-federated-learning
[5] What’s the Difference Between GraphQL and REST?, AWS. [Online]. Available: https://aws.amazon.com/compare/the-difference between-graphql-and-rest/
[6] Leon Witt et al., “Decentral and Incentivized Federated Learning Frameworks: A Systematic Literature Review,” IEEE Internet of Things Journal, vol. 10, no. 4, pp. 3642-3663, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Xiaoyuan Liu et al., “Unified: A Benchmark for Federated Learning Frameworks,” Arxiv, pp. 1-13, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Cross-Silo and Cross-Device Federated Learning on Google Cloud, 2024. [Online]. Available: https://cloud.google.com/architecture/cross-silo-cross-device-federated-learning-google-cloud
[9] GraphQL, What is Federated Architecture?. [Online]. Available: https://graphql.com/learn/federated-architecture/
[10] Kewei Cheng et al., “SecureBoost: A Lossless Federated Learning Framework,” IEEE Intelligent Systems, vol. 36, no. 6, pp. 87-98, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Matheus Seabra, Marcos Felipe Nazário, and Gustavo Pinto, “REST or GraphQL? A Performance Comparative Study,” Proceedings of the XIII Brazilian Symposium on Software Components, Architectures, and Reuse, Salvador Brazil, pp. 123-132, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Sin Kit Lo et al., “Architectural Patterns for the Design of Federated Learning Systems,” Journal of Systems and Software, vol. 191, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Hongyi Zhang, Jan Bosch, and Helena Holmström Olsson, “Federated Learning Systems: Architecture Alternatives,” 2020 27th Asia Pacific Software Engineering Conference, Singapore, Singapore, pp. 385-394, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[14] NVIDIA Clara Train 3.1, Federated Learning Background and Architecture. [Online]. Available: https://docs.nvidia.com/clara/clara train-archive/3.1/federated-learning/fl_background_and_arch.html
[15] Why GraphQL Needs an Open Federation Approach, TheNewstack. [Online]. Available: https://thenewstack.io/why-graphql-needs-an open-federation-approach/
[16] Understanding API Technologies: A Comparative Analysis of REST, GraphQL, and Asynchronous APIs, DZone. [Online]. Available: https://dzone.com/articles/understand-api-technologies-comparative-analysis
[17] Vijay Anand Rajasekaran et al., “Architectural Patterns for the Design of Federated Learning Systems,” Model Optimization Methods for Efficient and Edge AI: Federated Learning Architectures, Frameworks and Applications, pp. 223-239, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Saneev Kumar Das, and SujitBebortta, “Heralding the Future of Federated Learning Framework: Architecture, Tools and Future Directions,” 2021 11th International Conference on Cloud Computing, Data Science & Engineering, Noida, India, pp. 698-703, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Patrick Foley et al., “OpenFL: The Open Federated Learning Library,” Physics in Medicine & Biology, vol. 67, no. 21, pp. 1-11, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Sayan Guha, and Shreyasi Majumder, “A Comparative Study between Graph-QL & Restful Services in API Management of Stateless Architectures,” International Journal on Web Service Computing, vol. 11, no. 2, pp. 1-16, 2020.
[Google Scholar] [Publisher Link]
[21] Simo Öysti, “REST vs. GraphQL–Building APIs for Abstract Applications,” Thesis, Jyväskylä University of Applied Sciences, pp. 1-70, 2021.
[Google Scholar] [Publisher Link]