Dear Eddy experts,

I am wondering if it is possible to run a single eddy_cuda (version 5.0.11) process using multiple GPUs across multiple nodes (e.g., in a HPC machine).