Writing scalable SIMD programs with ISPC
James C. Brodman, Dmitry Babokin, Ilia Filippov, Peng Tu · 2014
Modern processors contain many resources for parallel execution. In addition to having multiple cores, processors can also contain vector functional units that are capable of performing a single operation on multiple inputs in parallel. Taking advantage of this vector hardware is essential to obtaining peak performance on a machine, but it is often challenging for programmers to do so.