This chapter lays the foundation by covering core concepts, including terminology, that are critical to have fresh in our minds as we learn how to accelerate C++ programs using data parallelism.Data parallelism in C++ enables access to parallel resources in a modern heterogeneous system. A single C++ application can use any combination of devices—including GPUs, CPUs, FPGAs, and AI Application-Specific Integrated Circuits (ASICs)—that are suitable to the problems at hand.This book teaches data-parallel programming using C++ and SYCL.SYCL (pronounced sickle) is an industry-driven Khronos standard that adds data parallelism to C++ for heterogeneous systems. SYCL programs perform best when paired with SYCL-aware C++ compilers such as the open source Data Parallel C++ (DPC++) compiler used in this book. SYCL is not an acronym; SYCL is simply a name. PC++ is an open source compiler project, initially created by Intel employees, committed to strong support of data parallelism in C++. The DPC++ compiler is based on SYCL, a few extensions,1 and broad heterogeneous support that includes GPU, CPU, and FPGA devices. In addition to the open source version of DPC++, there are commercial versions available in Intel oneAPI toolkits.Implemented features based on SYCL are supported by both the open source and commercial versions of the DPC++ compilers. All examples in this book compile and work with either version of the DPC++ compiler, and almost all will compile with recent SYCL compilers. We are careful to note where extensions are used that are DPC++ specific at the time of publication.