Power and thermal constraints have caused a paradigm shift in the semiconductor industry over the past few years. All market segments, including phones, tablets, desktops, and servers, have now reduced their emphasis on clock frequency and shifted to multicore architectures for boosting performance. Figure 1.1 clearly shows this trend of saturating frequency and increasing core count in modern processors. With Moore’s Law, on-die integration of many components such as peripheral control hubs, dynamic random-access memory (DRAM) controllers, modems, and more importantly graphics processors has become possible. Single-chip integration of graphics processing units (GPUs) with central processing units (CPUs) has emerged and also brought many challenges that arise from integrating disparate devices/architectures, starting from overall system architecture, software tools, programming and memory models, interconnect design, power and performance, transistor requirements, and process-related constraints. This chapter provides insight into the implementation, benefits and problems, current solutions, and future challenges of systems having CPUs and GPUs on the same chip.