simple_parallel exists and is pretty neat, yes. Rayon is another cool library here (https://github.com/nikomatsakis/rayon), which lets you do foreach/mapreduce operations via work stealing on a pool with an extensible API that looks exactly like the regular iterator api (https://github.com/nikomatsakis/rayon#parallel-iterators, for example).

Note that a simd library exists which abstracts over the intrinsics (https://github.com/huonw/simd)