ORBITER: A Framework for Automatic Deployment of Big Data Applications in Serverless Architectures
Abstract
It has become increasingly common to offer serverless computing mechanisms in public clouds. Although the development of an application following a serverless architecture presents advantages, it still brings challenges such as low portability. In addition, the serverless architecture also ends up adding a certain level of delay in building the architecture of distributed applications, which is already complex. This paper proposes a framework for deploying big data applications in a serverless architecture. The proposed framework was developed using open source tools and was evaluated with an ANTT traffic accident data analysis application.
References
Behm, A., Palkar, S., et al. (2022). Photon: A fast query engine for lakehouse systems. SIGMOD ’22, page 2326-2339, New York, NY, USA. ACM.
de Carvalho, L. R. and de Araujo, A. P. F. (2020). Performance comparison of terraform and cloudify as multicloud orchestrators. In CCGRID, pages 380-389. IEEE.
Hassan, H. B., Barakat, S. A., and Sarhan, Q. I. (2021). Survey on serverless computing. J. Cloud Comput., 10(1):39.
Hellerstein, J. M., Faleiro, J. M., et al. (2019). Serverless computing: One step forward, two steps back. In CIDR. https://www.cidrdb.org.
Mampage, A., Karunasekera, S., and Buyya, R. (2021). A holistic view on resource management in serverless computing environments: Taxonomy, and future directions. CoRR, abs/2105.11592.
Nandury, S. V. and Begum, B. A. (2016). Strategies to handle big data for traffic management in smart cities. In ICACCI 2016, India, pages 356-364. IEEE.
Perron, M., Fernandez, R. C., DeWitt, D. J., and Madden, S. (2020). Starling: A scalable query engine on cloud functions. In SIGMOD, June 14-19, 2020, pages 131-141. ACM.
Sousa, F. (2020). Computação serverless e gerenciamento de dados. In Anais do XXXV Simpósio Brasileiro de Bancos de Dados, pages 199-204, Porto Alegre, RS, Brasil. SBC.
Wang, A., Zhang, J., et al. (2020). Infinicache: Exploiting ephemeral serverless functions to build a cost-effective memory cache. In Noh, S. H. and Welch, B., editors, USENIX FAST, pages 267-281. USENIX Association.
