Distributed Processing Network

Abstract

Large medical computer systems can be visualized and designed in at least two practical configurations. Using a “central” concept predicates a large host computer in which a single data base is maintained for interaction with a number of terminal users (Fig. 1) [1, 2]. In this configuration, the major expense is the central system itself, and the speed and access to the files are dependent on the data-base managing software. The obvious limitations of this design relate to the lack of failsafe redundancy and the high cost of the initial investment.