When will it support three-dimensional data processing?

[duanfan07]

I believe Sedona already provides excellent support for two-dimensional data processing. Currently, we need to handle three-dimensional data. Since we have already integrated the Sedona technology stack and aim to leverage its distributed computing capabilities for processing this three-dimensional data, we hope to prioritize adding support for three-dimensional data processing.

[jiayuasu]

What kind of 3D operations are you looking for?

[duanfan07]

I am currently engaged in the development of a digital twin and 3D scene system within the power industry. At present, many of our 3D scenes cannot be rendered in real time due to their high computational demands. Therefore, we aim to leverage Sedona's distributed processing capabilities to efficiently handle various types of 3D data. Specifically, we want to uniformly convert multiple 3D data formats—such as TIF, LAS, OSGB, DAE, OBJ, and GIM—into 3D Tiles. Subsequently, similar to how we handle 2D data, we intend to partition the unified 3D data, index each partition, and then perform associative analysis. In summary, the goal is to standardize diverse 3D data formats into a unified format, partition this standardized 3D data, index the partitions, and conduct analytical operations.

As far as I know, Sedona is developed based on foundational libraries such as JTS and GeoTools, which have very limited support for 3D data. Therefore, we hope to prioritize and enhance 3D capabilities within Sedona.

[jiayuasu]

@duanfan07 thanks for the great feedback. Do you have an existing single machine library in mind? Maybe we can take a look at its data model and query operators and see if we can scale it to hundreds of machines.