Multicore - how to combine the local data groups found on each core?

I have a large set of scalar values ​​distributed over a 3D grid (one value for each vertex).

My goal is to show:

  • all points in the grid where the value is greater than the threshold value.
  • And group the points that are connected (to simplify the display.)

So my main solution was:

  • Find points that pass the threshold test
  • For each point that has not been grouped, create a new group and recursively place all the connected points in this group.

This works fine until I started using a multi-core solution:

  • The data set was divided into several cores
  • Each core is aware of the boundary points that are shared by other kernels.
  • I use MPI for communication between kernels.

, "" .

- "" . : . , , .

.

+3

Source: https://habr.com/ru/post/1710778/


All Articles