For decades, the computer industry has sought answers to a problem that is inherent to the architecture used in computing systems. The bottleneck created by keeping data separate from the elements of processing and control has come to be a seriously limiting factor in attempts to increase the throughput of systems. Numerous attempts to get around the problem have increased throughput by modest amounts only as they fail to address the real source of the problem which lies in the design used for systems. This book offers a new approach, a novel approach to the problem. Instead of offering an increasing concentration of power and resources at central points within the system, it does exactly the opposite. Processing, control, storage, and I/O are thoroughly integrated into single units of functionality that are used to build the entire system so that there are no longer any central points of computation. There are no CPUs, but there is a multitude of decentralized processing units. This completely decentralized distribution of computing resources is then apportioned among numerous processes which all function simultaneously and independently within the system. The result is a vast potential for increased throughput in the system. In addition, the system is stable, secure, expandable, and lends itself well to modular problem solving techniques.