[t.oliveira]

Hi everyone,

Is it possible (and if it isn't, would it make sense) to run utilities with the option -parallel, but in serial or in a different number of processors?

Let me illustrate with the example of a use case. I am preparing a model to be run in 432 processes on a cluster (on 432 cores). The cluster has a queue of some hours. I wanted to run renumberMesh -parallel before running the model. Ideally, I would be able to do it on my desktop, either in serial or using its 12 cores. The operation wouldn't take too long and I would avoid the queue.

Is it conceptually possible to implement that? Is it implemented already?

Thank you for your help,
Thomas Oliveira