Using OpenFaaS on Embedded and Server Platforms

• Igor C. D. de Almeida UNIOESTE
• Guilherme Galante UNIOESTE
• Marcio S. Oyamada UNIOESTE

DOI: https://doi.org/10.5753/eradrs.2025.6748

Abstract

This article compares the performance of cloud and edge computing using the serverless paradigm with OpenFaaS. A function for prime number verification was implemented and executed both on a server and on an embedded platform (S905x, ARM A53). The use of a unified serverless solution simplifies the client's request, requiring only the selection of the execution location. The results indicate that the edge is feasible for light workloads but becomes a bottleneck for complex calculations. The study highlights the need for dynamic balancing between cloud and edge to optimize performance and scalability.

Keywords: Performance Evaluation, Measurement, and Prediction, Cluster Computing, Heterogeneous Computing, High Performance Computing in Grids and the Cloud, Scheduling and Load Balancing

References

Hong, X. e Wang, Y. (2018) "Edge Computing Technology: Development and Countermeasures", Strategic Study of CAE, 20(2): 20–26. Disponível em: [link]

Documentação OpenFaaS (2025). Disponível em: [link]

Shafiei, H; Khonsari, A. e Mousavi, P. (2022) Serverless Computing: A Survey of Opportunities, Challenges, and Applications. ACM Comput. Disponível em: DOI: 10.1145/3510611

Agudelo-Sanabria, S. D.; Jindal, A. (2021) The ifs and buts of the development approaches for iot applications. Disponível em: [link].

Cloud Native Computing Foundation. (2017) CNCF WG-Serverless Whitepaper v1.0. Disponível em: [link]

Xie, R., Tang, Q., Qiao, S., Zhu, H., Yu, F. R. e Huang, T. (2021) "When Serverless Computing Meets Edge Computing: Architecture, Challenges, and Open Issues", IEEE Wireless Communications, 28(5): 40-47. Disponível em: [link].