Featuring the Intel® Advisor tool, this video demonstrates the increases in speed and decreases in loop trip counts that can be gained by vectorizing code, examines how data type sizes can affect these results, and shows how to tweak them for maximal gains. The demo compares the trip counts and run times of the scalar and vectorized versions of a pair of loops to show the massive improvement gained by vectorization … and the consequences of using more bytes than you need. It also shows how to use aspects of Intel Advisor’s Survey Report to gain diagnostic information about the vectorized loops.
Advantages of Vectorization and the Effects of Data Size
Alex Shinsel, Technical Consulting Engineer, Intel Corporation
Alex Shinsel, a Technical Consulting Engineer for Intel since 2016, focuses on software optimization using performance analysis tools, such as Intel® VTune™ Amplifier and Intel® Advisor. She loves solving challenging problems and helping customers learn how to solve theirs. Alex graduated with honors from Pacific University in Oregon with a B.S. in Computer Science.