GeoFlink: A Distributed and Scalable Framework for the Real-time Processing of Spatial Streams - Githubissues

Sunt-ing / database-system-readings

:yum: A curated reading list about database systems

MIT License

466 stars 31 forks source link

GeoFlink: A Distributed and Scalable Framework for the Real-time Processing of Spatial Streams #28

Closed Sunt-ing closed 2 years ago

Sunt-ing commented 2 years ago

https://raw.githubusercontent.com/Sunt-ing/database-system-readings/main/papers/Geoflink.pdf

Sunt-ing commented 2 years ago

What problem does the paper solve?

There is no stream geospatial big data system that can support streaming polygon objects.
Queries: spatial range, spatial kNN, spatial join. All are continuous queries only.

How does it solve the problem?

Build a spatiotemporal system on Flink using grid index.

What could be improved?

Adding consistency guarantee.
Automatical load balance
Supporting snapshot query
Supporting trajectory workload
The architecture illustrations are not in the SVG form, making them vague after being scaled. Bad example. However, the statistics are in the SVG form.

Others

To avoid being tedious, a good paper should not be a document of the project. Unfortunately, this paper is somewhat like a document, partly because it's lacking of novelties.
It's amazing that the project is well-maintained for a year after the paper was published.