When you read about OpenMP, you typically see examples that demonstrate its use for parallelizing loops. Many performance engineers focus on loops because they are hot spots—that is, places where programs spend much of its execution time. And because loops consist of code sequences that are repeated sometimes thousands of times, a small optimization in loop processing can have substantial performance benefits.

But there's more to parallelism than loop optimization! Programs are not all loops, and other constructs benefit from the ability to leverage parallel processing via OpenMP. This article discusses OpenMP features that enhance the performance of straight ahead program code.