Each time your models sends a data file to another node of the network, record the size of the file which is communicated (in kilo bytes) using cumulator.data_transferred(file_size).
I read that as saying that it measures the cost of sending data from one computing node of a HPC cluster to another computing node of the same cluster, which is not at all the same as downloading a file from the internet. This needs to be made much clearer.
The README currently says this:
I read that as saying that it measures the cost of sending data from one computing node of a HPC cluster to another computing node of the same cluster, which is not at all the same as downloading a file from the internet. This needs to be made much clearer.