Theme 1: Optimizing Compiler for KLIC
Various ideas for the optimization of implementations of parallel logic programming languages are proposed. Evaluation was made through building of experimental systems.
The proposed ideas seem promising and the construction of the experimental systems appears satisfactory. Meaningful results appear to have been built. However, the actual usefulness of most of the proposed optimization ideas have yet to be proven within the framework of the KLIC system.
For the second year, we feel it would be desirable to integrate some of the proposed optimization methods into KLIC. To achieve this, a streamlining of the organization of the research activities and a more frequent exchange of ideas and software between the University of Oregon and ICOT will be necessary.
To clarify the above-mentioned goal, we propose adding a sentence to Article 2.1 of Attachment A, such that the article reads:
The first research theme is optimization techniques for a compiler which translates KL1 programs to C programs. Research is conducted into the building and evaluation of software tools for analyzing KL1 programs and generating the information required for optimized compilation. The results of this research will be a report on basic theories and implementation methods.
Theme 2: Analysis of Biological Sequences with Parallel Inference Systems
From the results of the first year, the research target has been clarified. We are particularly interested in the idea of a software tool for analyzing evolution relationships by combining a secondary structure prediction system with a multiple alignment system. The technical problems to be overcome in realizing the software tool have already been identified.
Experiments aimed at describing the structural features of tRNA as constraints have been conducted, and it has become clear that structural constraints are useful for analyzing tRNA sequences.
We feel that two main topics should form the focus of the second year's investigations.
As the constraints of tRNA are too specific, there is a need to devise a more general description of the structures of RNA.
We also see a need to verify the effectiveness of the description of structural constraints by developing a software tool and by confirming its output.
We would, however, like to leave Article 2.2 of Attachment A as is, as we believe that the current description covers the aspects mentioned above.