Summing an array in serial C is a straightforward accumulation loop:

double sum = 0.0;
for (int i = 0; i < N; i++)
    sum += a[i];

Adding #pragma omp parallel for to this loop creates a race condition: multiple threads read sum, add to it, and write back simultaneously, so some updates are lost. The reduction clause fixes this.

Reduction in OpenMP gives each thread a private copy of the accumulation variable, lets it accumulate locally without contention, then merges all copies at the end of the loop using the specified operator.

double sum = 0.0;
 
#pragma omp parallel for reduction(+:sum)
for (int i = 0; i < N; i++) {
    sum += a[i];
}

Built-in reduction operators include +, *, -, min, max, and the bitwise operators &, |, ^. Custom reducers are possible in C++ via declare reduction.