We describe a simple yet highly parallel method for re-indexing "indexed" data sets like triangle meshes or unstructured-mesh data sets -- which is useful for operations such as removing duplicate or un-used vertices, merging different meshes, etc. In particlar, our method is parallel and GPU-friendly in the sense that it all its steps are either trivially parallel, or use GPU-parallel primitives like sorting, prefix-sum; thus making it well suited for highly parallel architectures like GPUs.