Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a comprehensive talk from C++Now! 2012 on HPX, a C++11 parallel runtime system. Delve into the challenges faced by computational science applications due to limitations of traditional execution models like MPI. Learn how HPX addresses scaling issues in parallel computing, particularly in the context of increasing power demands, processor core complexity, multi-core sockets, and heterogeneous GPU structures. Discover the key features of HPX, including its modular design, implementation of the ParalleX execution model, and incorporation of lightweight user-threads and an Active Global Address Space (AGAS). Gain insights into HPX's C++11 implementation and its utilization of over 20 Boost and candidate Boost libraries. Understand how HPX offers an alternative to MPI for parallel computing applications, potentially overcoming scaling limitations in various computational science domains.