Building a scalable geofencing API on Google’s App Engine

Thorsten Schaeff outlines the process of building a scalable geofencing API on Google's App Engine, emphasizing the challenges of working with geospatial data in a cloud environment and the benefits of using Google's Cloud Platform. He details the architecture of this API, which involves storing complex polygons and metadata in Cloud Datastore, indexing them for fast querying using an STR-Tree stored in memcache, and serving results through HTTP requests using Google Cloud Endpoints. Additionally, the implementation utilizes the Java Topology Suite for geometric computations and the GSON library for JSON handling. Schaeff provides a step-by-step guide on setting up the project with Apache Maven, adding libraries, creating endpoints for adding, listing, and retrieving fence metadata, and building a spatial index for efficient geofencing queries. The API is designed for automatic scalability and cost-effectiveness, with options to enhance performance through dedicated memcache. The post also hints at future work on processing large spatial datasets using Google's Cloud Dataflow.