# An Empirical Evaluation of Flow Based Programming in the Machine Learning Deployment Context

arXiv: https://arxiv.org/abs/2204.12781 · HTML: https://mlsanity.com/paper/2204.12781/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 55 references that are in the map.
- First posted on arXiv 27 Apr 2022.

## Builds on (most cited first, 8 of 8)

- [TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems](https://mlsanity.com/paper/1603.04467/) (2016)
- [Challenges in Deploying Machine Learning: a Survey of Case Studies](https://mlsanity.com/paper/2011.09926/) (2020)
- [MLlib: Machine Learning in Apache Spark](https://mlsanity.com/paper/1505.06807/) (2015)
- [Continual Learning in Practice](https://mlsanity.com/paper/1903.05202/) (2019)
- [Data Engineering for Data Analytics: A Classification of the Issues, and Case Studies](https://mlsanity.com/paper/2004.12929/) (2020)
- [Towards better data discovery and collection with flow-based programming](https://mlsanity.com/paper/2108.04105/) (2021)
- [Data-driven Workflows for Microservices](https://mlsanity.com/paper/1511.02597/) (2015)
- [Control Flow Versus Data Flow in Distributed Systems Integration: Revival of Flow-Based Programming for the Industrial Internet of Things](https://mlsanity.com/paper/2108.08081/) (2021)

## Built on by (most cited first, 5 of 5)

- [Desiderata for next generation of ML model serving](https://mlsanity.com/paper/2210.14665/) (2022)
- [Dataflow graphs as complete causal graphs](https://mlsanity.com/paper/2303.09552/) (2023)
- [Machine Learning Systems: A Survey from a Data-Oriented Perspective](https://mlsanity.com/paper/2302.04810/) (2023)
- [What About the Data? A Mapping Study on Data Engineering for AI Systems](https://mlsanity.com/paper/2402.05156/) (2024)
- [Causal fault localisation in dataflow systems](https://mlsanity.com/paper/2304.11987/) (2023)

## Abstract (arXiv)

As use of data driven technologies spreads, software engineers are more often faced with the task of solving a business problem using data-driven methods such as machine learning (ML) algorithms. Deployment of ML within large software systems brings new challenges that are not addressed by standard engineering practices and as a result businesses observe high rate of ML deployment project failures. Data Oriented Architecture (DOA) is an emerging approach that can support data scientists and software developers when addressing such challenges. However, there is a lack of clarity about how DOA systems should be implemented in practice. This paper proposes to consider Flow-Based Programming (FBP) as a paradigm for creating DOA applications. We empirically evaluate FBP in the context of ML deployment on four applications that represent typical data science projects. We use Service Oriented Architecture (SOA) as a baseline for comparison. Evaluation is done with respect to different application domains, ML deployment stages, and code quality metrics. Results reveal that FBP is a suitable paradigm for data collection and data science tasks, and is able to simplify data collection and discovery when compared with SOA. We discuss the advantages of FBP as well as the gaps that need to be addressed to increase FBP adoption as a standard design paradigm for DOA.