ICOT has carried out the Fifth Generation Computer Systems (FGCS) Project since April 1982. During this project, ICOT developed new parallel software technology based on parallel logic programming and produced many advanced software systems.
ICOT desires to distribute the software technology and software systems developed by the FGCS project. To achieve these goals, ICOT started the FGCS Follow-on Project in April 1993 as a two year national project. The Follow-on Project intends to make the software technology widely available as a common ground and infrastructure for advanced research by porting it to general purpose parallel machines which will be popular in the market.
To distribute the software technology more effectively, ICOT established the ICOT Free Software (IFS) program. Major software systems developed during the FGCS project have been released as IFS, and new software to be developed in the Follow-on Project is also expected to be released through IFS.
To promote the distribution and further development of IFS, ICOT desires the
University of Bristol to conduct research on parallel logic programming and
applications currently released as IFS.
As a general framework to determine themes and contents of the research being conducted under this agreement, ICOT and the University of Bristol have agreed to the following items:
The University of Bristol understands the desires of ICOT
and plans to conduct the research described below. A research group headed by
Professors David Warren and Steven Gregory will conduct research related to two
general themes.
Theme 1: Debugging Tools for KL1
Various useful debugging techniques have been developed for concurrent logic programming languages such as KL1, but none of them is completely satisfactory yet. The main difficulty is to locate ``non-deterministic bugs" that reveal themselves in some executions of a program but not in others. To overcome this difficulty, a debugger that explores all possible execution paths, taking account of the inherent non-determinism, is desirable. However, a naive implementation of such exploration would only result in an explosion of possible states, requiring too much memory for practical use.
In this research, ways to make such a debugger practical by reducing the search space are investigated, making use of A.I. search techniques. Ways to integrate the debugger into KL1 language systems are also investigated and evaluated. The results of this research will be a report on basic theories and implementation methods.
Theme 2: Parallel Constraint Logic Programming System
Committed-choice parallel logic programming languages like KL1 lack a capability of searching to make its language processors simpler and improve their efficiency. However, search is a very important facility in programming. Constraint solving on finite domains is one of the typical problems that require searching facility. Furthermore, introduction of constraints into logic programming languages makes them very powerful and flexible programming languages.
In this research, ways to make a parallel constraint logic programming language for finite domain constraints on top of KL1 are investigated. To reduce the search space, heuristics for this language are also investigated. The results of this research will be a report on basic theories and implementation methods for the constraint logic programming language.
The University of Bristol will provide a single report which summarizes the research activities carried out under both themes described in Section 2 and contains documents related to the design, evaluation and/or application of targeted software systems. As supplements to this report, the University of Bristol will furnish copies of major evaluation data and related technical information produced by the research under this agreement.
The report and the supplementary documents shall be provided to ICOT no later
than March 10, 1995.
The University of Bristol and ICOT will arrange close communication between researchers of both parties so that the research activities will be as desirable as possible for both parties.
Any matter or questions arising out of or relating to the research shall be settled in good faith by mutual agreement between both parties.