This module looks at basic optimization techniques for the Intel Xeon Phi coprocessor. It briefly reviews the standard optimization methodology; discusses how to use the compiler to instrument you code to collect performance data, get optimization data, and get optimization suggestions; Using loop profiler to identify opportunities for task and SIMD parallelism, and how to exploit the auto-vectorizer.

Section 3: GeneralIntel Xeon Phi coprocessor optimization advice, thread affinity and load balancing, a brief overview of prefetching and aligning memory