C++11 Multithreading
Abstract
During the last decade, single-core performance of modern CPUs has been stagnating due to hard architectural limitations of silicon-based semiconductors. On the one hand, the historic shrinking of the manufacturing process of three orders-of-magnitude from a few micrometers to a few nanometers cannot proceed at arbitrary pace in the future. This can be explained by emerging delocalization effects of quantum mechanics which governs physics at the nanometer scale. On the other hand, the power dissipation of integrated circuits exhibits a quadratic dependency on the voltage and a linear dependency on the frequency. Thus, significant up-scaling of the CPU's frequency is prohibitive in terms of energy ...
Get Parallel Programming now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
