Summary

In this chapter, we walked through a couple of different examples of how Athena allows you to inject custom functionality, known as user-defined functions, into your queries. We started by looking at fully custom UDF behavior through Lambdas. We created and deployed our own Lambda, and then took a closer look at how we can keep a healthy, well-maintained Lambda-based UDF. After that, we took a look at the built-in UDF functionality that Athena provides for integrating your queries with SageMaker ML models. We used this to determine if taxi ridership was anomalous during a specific time.

We've only scratched the surface of the power of UDFs, but this should serve as a solid reminder for when you encounter a business use case that you can't solve perfectly with the functionality provided out of the box. In the next and final chapter, we will summarize some advanced functions that Athena provides and conclude our book!