Chapter 6. Deploying to Production

Business leaders view the rapid deployment of new systems into production as key to maximizing business value. But this is only true if deployment can be done smoothly and at low risk (software deployment processes have become more automated and rigorous in recent years to address this inherent conflict). This chapter dives into the concepts and considerations when deploying machine learning models to production that impact—and indeed, drive—the way MLOps deployment processes are built (Figure 6-1 presents this phase in the context of the larger life cycle).

Figure 6-1. Deployment to production highlighted in the larger context of the ML project life cycle

CI/CD Pipelines

CI/CD is a common acronym for continuous integration and continuous delivery (or put more simply, deployment). The two form a modern philosophy of agile software development and a set of practices and tools to release applications more often and faster, while also better controlling quality and risk.

While these ideas are decades old and already used to various extents by software engineers, different people and organizations use certain terms in very different ways. Before digging into how CI/CD applies to machine learning workflows, it is essential to keep in mind that these concepts should be tools to serve the purpose of delivering quality fast, and the ...

Get Introducing MLOps now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.