I. Introduction
The design of digital control systems, both for industrial application and for R&D experiments, needs the definition of requirements on control band, sampling frequency, computing power, ADC and DAC characteristics. In many cases, the implementation of a control system becomes just a matter of choice of standards, technologies, control algorithms already described in literature or available on the market, often very reliable and effective. The design of non-standard control systems, like high performance control systems, adaptive control systems or distributed control systems is different. In this case the required large computing powers, distributed controlling actions, large control bands (of the order of kHz or larger) are not easily satisfied by commercial embedded or open systems, without using expensive and complex architectures (e.g., Digital Signal Processors (DSP) to be integrated in the acquisition and control system). The control system of the Interferometric Detector of Gravitational Waves Virgo [1], [2] is a typical example. In fact, this system is based on DSP VME boards designed and implemented to satisfy the specific requirements for computing power and data transfer of the Virgo detector control system. This solution, although very efficient, requires large investments, also in terms of manpower, for the design, implementation, test, maintenance and upgrade of the control system.