Skip to main content

Parallel For

Execute loop iterations concurrently across CPU threads.

Syntax​

parallel for (i is 0; i < data.Length; i++) {
    data[i] is data[i] * 2;
}

How It Works​

The parallel modifier before for distributes iterations across available CPU threads. Each iteration runs independently.

When to Use​

  • Large data transformations
  • Element-wise operations on arrays/tensors
  • Independent computations per iteration

Restrictions​

  • Iterations must be independent — no shared mutable state
  • Order of execution is not guaranteed

GPU Alternative​

For GPU-accelerated parallelism, use gpu { } blocks instead:

gpu {
    result is A + B;    // GPU-parallel tensor operation
}

See GPU Blocks.

Next Steps​