PYTHIA - A KNOWLEDGE-BASED SYSTEM TO SELECT SCIENTIFIC ALGORITHMS

Problem-solving Environments (PSEs) interact with the user in a langua ge ''natural'' to the associated discipline, and they provide a high-l evel abstraction of the underlying, computationally complex model. The knowledge-based system PYTHIA addresses the problem of (parameter, al gorithm) pair selection within a scientific computing domain assuming some minimum user-specified computational objectives and some characte ristics of the given problem. PYTHIA's framework and methodology are g eneral and applicable to any class of scientific problems and solvers. PYTHIA is applied in the context of Parallel ELLPACK where there are many alternatives for the numerical solution of elliptic partial diffe rential equations (PDEs). PYTHIA matches the characteristics of the gi ven problem with those of PDEs in an existing problem population and t hen uses performance profiles of the various solvers to select the app ropriate method given user-specified error and solution time bounds. T he profiles are automatically generated for each solver of the Paralle l ELLPACK library.