parallel processing

parallel processing the concurrent or simultaneous execution of two or more parts of a single computer program , at speeds far exceeding those of a conventional computer . Parallel processing requires two or more interconnected processors, each of which executes a portion of the task; some supercomputer parallel-processing systems have hundreds of thousands of microprocessors . The processors access data through shared memory. The efficiency of parallel processing is dependent upon the development of programming languages that optimize the division of the tasks among the processors.

Bibliography: See E. Rietman, Exploring Parallel Processing (1990); K. M. Chandy and S. Taylor, An Introduction to Parallel Programming (1992); D. I. Moldovon, Parallel Processing from Applications to Systems (1993); G. S. Almasi and A. Gottlieb, Highly Parallel Computing (1993).

parallel processing A term applied rather loosely to a number of rather similar concepts but with important detailed differences. The essence of parallel processing is that more than one particular process is active at any given instant; however the term is often applied to a situation in which a large number of processes are potentially active but at any one instant only one is active. Strictly speaking the term parallel processing should only be applied where more than one processor is active among a group of processes at any one instant. In practice it is seldom used with this accurate connotation. See also concurrent programming.

parallel processing A method of computing in which two or more parts of a program are executed simultaneously rather than sequentially. Strictly, parallel processing is only possible on computers with more than one arithmetic and logical unit in the central processing unit, but it is often simulated on other machines by such techniques as time sharing.