Guided Code Vectorization with Intel® Advisor XE

Early Stage Application Optimization made Easy with Step-by-Step Guide

In this publication we discuss the usage of an optimization tool called Intel^® Advisor. The discussion is illustrated with an example workload that computes the electric potential in a set of points in 3-D space produced by a group of charged particles. The example workload runs on a multi-core Intel Xeon processor with Intel AVX2 instructions.

The application was originally parallelized across cores, but otherwise neither optimized nor vectorized. In the publication, we discuss three performance issues that the Intel Advisor detected: vector dependence, type conversion and inefficient memory access pattern. For each issue, we discuss how to interpret the data presented by the Intel Advisor, and also how to to optimize the application to resolve these issues. After the optimization, we observed a 16x performance boost compared to the original, non-optimized implementation.

Complete paper:
 Colfax_Advisor_Vectorization.pdf (1 MB)

Sample code for Linux:
Colfax_Advisor_Vectorization_Code.zip (50 KB)