Essentials of Computer Organization and Architecture, 5th Edition

9.4   PARALLEL AND MULTIPROCESSOR ARCHITECTURES

Since the beginning of computing, scientists have endeavored to make machines solve problems better and faster. Miniaturization technology has resulted in improved circuitry, and more of it on a chip. Clocks have become faster, leading to CPUs in the gigahertz range. However, we know there are physical barriers that control the extent to which single-processor performance can be improved. Heat and electromagnetic interference limit chip transistor density. Even if (when?) these problems are solved, processor speeds will always be constrained by the speed of light. On top of these physical limitations, there are also economic limitations. At some point, the cost of making a processor incrementally ...