data parallelism

Doing the same work on many pieces of data at once, on several CPU cores.

Data parallelism splits a collection across threads; rayon's par_iter does it with a work-stealing pool and an iterator-like API.

Examples

With rayon, changing readings.iter() to readings.par_iter() in readings.par_iter().map(|r| r * 10).sum::<u64>() spreads the work over the CPU cores and gives the same sum.