In Chapter 9, Threads, we covered how to use the experimental threading functionality to utilize multiple processors within a single processor. Some embedded libraries used in Julia, such as OpenBLAS, use their own threading facilities to scale across multiple CPUs. Julia's asynchronous IO libraries can also offload network or file IO to a separate operating system thread. Notwithstanding these exceptions, most Julia code continues to be limited to a single core.
Julia, however, contains an easy-to-use multiprocessor mechanism. You can start multiple Julia processes, either on a single host, or across a network, and control, communicate, and execute programs across the entire cluster.
