In the beginning of December, I attended UKOUG Annual Conference DISCOVER 2024 in Birmingham. I was presenting there and I am said (or perhaps not) to share this information, but this was my last time I was presenting under Version 1 / Qubix umbrella. Our paths are finally splitting at the end of this year and I will share more in my future posts.
Back to my presentation in Birmingham.
Depending on the existing / future architecture OCI users have various option to choose from. The question is which ML approach is the most suitable for a specific user? And among these, some could be quite interesting. For example, which solution could be the best fit if customer still prefers to stay on-premise or what if data is fluent. So, what options have we got in OCI?
In my presentation titled 3 Case Studies: Machine Learning in OCI, I am trying to answer some of the questions mentioned about and I am discussing development and deployment options that we have in Oracle Cloud Infrastructure (OCI) based on three use cases we came across in the last couple of months.
First case study is about a manufacturing company that is using Oracle Autonomous Datawarehouse and Oracle Analytics Cloud for their analytics. Business requirement was to improve sales planning using machine learning. To fulfil this requirement, Oracle Machine Learning for Python (OML4Py) has been used.
In the second case study, client's infrastructure is mostly on-premise Oracle technology: Oracle Database, Oracle Analytics Server, Oracle Data Integrator, Hyperion Planning. This client has similar requirement (predictive planning) as the client in the first use case, however we have decided to implement machine learning solution using OCI Data Science. There are a few additional components in the solution: model training and predictions have been automated through Oracle Data Integrator process and by using models deployed to model catalogue that resides in OCI. At the end, final results are stored in Hyperion Planning application.
In our third case, we have gone one step further. We are dealing with streaming data which is channelled into OCI Golden Gate Stream Analytics where we use machine learning model, developed in OCI Data Science, in real-time. This project is still underway, however client plans to exploit this solution for quite an array of potential data streams.
What have we learnt?
Implementation choices: when implementing machine learning solutions in OCI, customer have wide array of choices about which technology is most suitable for them.
Implementation strategies: when implementing machine learning solutions, the basic life cycle is more or less the same for every project, however deployed technologies may vary quite considerable, depending on general infrastructure available (i.e. on-premise) or cost of a new services (costs of some OCI based services might be a surprise).
Here are the slides 3 Case Studies: Machine Learning in OCI.