Runtime scaling is not linear for a large lattice

[chaeyeunpark]

Simulation results up to 127x127x127 lattice show that the scaling is not linear. Need to profile (does the greedy implementation of the minimum spanning tree slow down it?)

[Qi-Zhan]

why is it necessary to implement minimum spanning tree, from the paper https://arxiv.org/abs/1703.01517 peeling algorithm only need spanning tree, which can be implemented in O(n). Maybe I misunderstand your needs for MST?

[chaeyeunpark]

Hi @Qi-Zhan, thank you for the question! I don't know why I wrote that. The current algorithm the implementation uses is O(n). Still, there might be a slight deviation from the linear scaling if the size of the lattice becomes too huge. It may be from a hash map implementation rather than the algorithm side.

[Qi-Zhan]

Hi, thank you for answer. I guess it is beacause of some implementation, too.