System Architecture and Complexity

Introduction to Part 2

Originally, at the time of the SAGE (Semi-Automatic Ground Equipment) and NTDS (Naval Tactical Data System) projects, the first systems were entities that were isolated from each other with no connections other than those created by human operators who could switch from one to the other as required. This situation lasted until the 1980s and the appearance of the first specific data networks developed by computer constructors and public organizations such as the CCITT/UIT to set up the first communication standards such as the layered ISO–OSI model. Information was then able to circulate in increasingly massive quantities in real time between systems.

The 1990s was a decade of disruption caused by the VLSI revolution which allowed small, powerful, energy-efficient machines to be released onto the market, which progressively but rapidly became substitutes for the mainframes that were symbols of centralized systems and which blended into what we call Clouds or even data centers. This processing power allowed real-time data compression algorithms to be implemented, which unexpectedly improved the flow of networks that were already undergoing great improvements due to the implementation of new technologies such as fiber optics.

Everything then came together, all interconnected; “passive” terminals gained properties of “intelligence”, in other words became true computers, autonomous, with software suitable for MMIs but linked to networks of servers in architectures ...