Is Morton most effective for higher measurements?

Question

For my current input, which are points in 3D, I use Morton code to improve cache coherency when accessing a point list.

I have other data that are 6D and 7D. Is Morton's code still a good technique for such measurements? Or are there other methods? Other space curve filling methods were harder to calculate than Morton in 3D itself, I wonder if people use an alternative method for 6D / 7D or higher.

Answer 1

You should try indexing rows or main rows. They also maintain spatial locality, but they can be calculated more efficiently, even in higher dimensions.

You can read about indexing rows and main columns in more detail (but in a less geometric sense) in the book “Assembly Language,” chapter 5, pages 211-216. The relevant chapter is available online here .

And there is a good article about the various spatial indexing methods that you can consider, including those mentioned: Samet, H. 2017. Sorting Spatial data. International Encyclopedia of Geography. 1-11.

, , Morton ( Morton). Morton ( ), .