Ensure that we maintain our parallel processing pipeline:
https://stackoverflow.com/questions/20004935/best-way-to-execute-parallel-processing-in-node-js
Since user data files are .geojson, this should work well; we should be able to use @Turfjs for better spatially based clustering. This will be really, really cool.
https://deniscarriere.github.io/turf-example-clusters-dbscan/
Port & Reprocess with Node?
/ Rather than reprocess existing data, create a better clustering method with javascript.