|Please refer to here which supplements this explanation by figures.|
The fifth generation computer system is a new computer and its
related technologies obtained by combination of two representative new
technologies of computer science:
These computer technologies will enhance knowledge information processing ability that will be one of the most important applications of future computers, and will utilize rapidly progressing semi-conductor technologies sufficiently.
What we call the fifth generation computer is a fifth generation computer proto-type system obtained by combination of software and hardwere technologies researched and developed along the above mentioned principle.
The institute for New Generation Computer Technology (ICOT) was established in April, 1982 as the organization responsible for conducting the Fifth Generation Computer Systems (FGCS) Project.
This project aimed at the development of a new computer that would feature an inference mechanism using knowledge bases for its kernel function, and which would make full use of large-scale parallel processing technology in its implementation.
After an eleven-year research and development effort, the FGCS Project achieved its initial goal and was brought to a successful conclusion in March 1993.
The project produced the FGCS prototype and established FGCS technology. This prototype is the world's fastest and largest-scale computer system for knowledge information processing. It included a large-scale hardware system called parallel inference machines (PIM's) such as PIM/p having 512 element processors and PIM/m having 256 element processors.
On these PIM's, a very productive parallel programming environment for the KL1 parallel logic language was built. It provides us with new and powerful debugging and monitoring tools such as run-time monitors dedicated to PIM/p and PIM/m.
Using this programming environment, various new application software systems in knowledge processing have been developed. Most of these software systems could make full of the PIM's and attain the speed-up almost proportional to the number of element processors. For example, the parallel theorem prover, MGTP attained the speed-up of 220 times using 256 element processors.
With the completion of the FGCS Project, the two-year FGCS Follow-on Project was started to disseminate the FGCS technology. This new project aims to port the KL1 programming environment and also major software systems developed in the FGCS project to many Unix-based sequential and parallel machines so as to make the FGCS technology a new research infrastructure for advanced computer development.
As for research results in the Follow-on Project, please refer to this page.