Cloudmesh Cloud AI Service

  • This project can also be taken as an independent study or as part of the Cloud Engineering class

Goals

In this project, you will be developing a number of services exposing AI capabilities on the cloud. You will be developing a REST service that invokes an AI functionality in a general term. It is not sufficient to just port an application, but you must decide which functions are part of the application and port these as collaborating REST services.

Your service must be demonstrated on Kubernetes and OpenStack showcasing applicability in a container and virtual machines. A deployment, testing, and benchmark activity need t to be demonstrated.

Requirements

Basic knowledge of python or a programming language. Basic knowledge of AI. The services will be developed in OpenAPI which you will learn throughout the semester. Flask is used as the Web service framework.

Difficulty

moderate: Although this project looks initially easy, it is to be reminded that multiple services need to be implemented while the functionality must be sufficiently fine-grained for reusability. Furthermore, data management for these AI services needs to be addressed.

A good level of fine-grained activities is showcased in scikit learn

Deliverables

  • Working benchmarks in form of reproducible pytests. Examples, showcasing the functionality.

  • checkins into GitHub by all team members. If only one team member checks in all other team members will not get credit.

  • Collaborate with all teams that do this project so we avoid implementation of duplicated fine-grained functionality.