Ivan's wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.

# Parallel loops (OpenMP)
The most common parallel pattern in C is a loop over independent array elements: each iteration writes to a different output location and doesn't read results that other iterations produce.

```c
for (int i = 0; i < N; i++)
    a[i] = i * 0.5;
```

**Parallel loops** in OpenMP parallelise exactly this pattern. Adding `#pragma omp parallel for` above the loop splits the iteration range across all threads, each thread writing to its own slice of the array. No synchronisation is needed because the iterations are independent.

```c
#pragma omp parallel for
for (int i = 0; i < N; i++) {
    a[i] = i * 0.5;
}
```

The directive requires the loop to have a countable iteration space and no data dependencies between iterations. Writing to `a[i]` must not depend on `a[i-1]` or any other `a[j]`. Loops with such dependencies require a different strategy (e.g. a scan) or explicit synchronisation.