Add Multi-GPU Functionality

Kindly add the functionality to do the integration using multi-GPU platforms (on a single compute node) and/or distributed across multiple nodes of a server.

Kindly also add some relevant examples.

Thank you

The issue requests multi-GPU support on one compute node or across multiple server nodes, plus relevant examples, but names no files, tests, or entry points. Start by surveying the repository structure and existing execution model to determine where such integration would belong. Done would require agreed support for the requested platforms and working examples.

Written by the indexing model from the issue text.