From the course: AI Solution Design Patterns: Data, Model Training, and Application Architectures

Unlock this course with a free trial

Join today to access over 24,300 courses taught by industry experts.

Bringing it all together: AI application architecture

Bringing it all together: AI application architecture

From the course: AI Solution Design Patterns: Data, Model Training, and Application Architectures

Start my 1-month free trial Buy for my team

Bringing it all together: AI application architecture

“

- So let's now take a look at a sample application architecture involving a service orchestration platform, microservices, containers, and AI system modules. Here we have an AI solution comprised of an AI system with its core modules implemented within containers, along with a set of microservices within the AI solution boundary, but outside of the core AI system. The microservices and modules interact with each other as well as with other solution components such as databases and monitors. This interaction scenario can be mapped out in a workflow definition and then executed by a service orchestration platform. Just a quick note that in this scenario, the databases were not placed in containers, although they commonly would be. There are a few possible reasons for this. They may be legacy databases that are not compatible with modern container environments or are simply never going to move anywhere from where they already are. Or they might be very large databases that have high…

Contents