Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a comprehensive conference talk on SIMD (Single Instruction, Multiple Data) capabilities in modern computing hardware. Gain a high-level overview of SSE, AVX, and AVX-512 instruction set architectures provided by Intel microprocessors. Learn how to compose a simple API using C++ and compiler intrinsics to implement these instruction sets. Discover real-world examples where additional performance can be achieved through "vertical" thinking. Examine how to improve performance for interesting problems like sorting values in a register, creating fast 1-D median filters, and high-speed convolution and correlation. Dive into the potential of SIMD computing and learn how to leverage its power in your C++ projects.