Pythia

Abstract

Problem-solving Environments (PSEs) interact with the user in a language "natural" to the associated discipline, and they provide a high-level abstraction of the underlying, computationally complex model. The knowledge-based system PYTHIA addresses the problem of (parameter, algorithm) pair selection within a scientific computing domain assuming some minimum user-specified computational objectives and some characteristics of the given problem. PYTHIA's framework and methodology are general 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 differential equations (PDEs). PYTHIA matches the characteristics of the given problem with those of PDEs in an existing problem population and then uses performance profiles of the various solvers to select the appropriate method given user-specified error and solution time bounds. The profiles are automatically generated for each solver of the Parallel ELLPACK library.

Pythia-Ii

“…In addition, CBR systems can exploit a priori domain knowledge to perform more sophisticated analyses even if pertinent data are not present. The original PYTHIA system utilized a rudimentary form of case-based reasoning employing a characteristic-vector representation for the problem population [Weerawarana et al 1997].…”

“…In Houstis et al [1991] we proposed an approach for dealing with this task by processing performance data obtained from testing software. The testing of this approach is described in Weerawarana et al [1997] using the PYTHIA implementation for a specific performance evaluation study. The approach has also been tested for numerical quadrature software [Ramakrishnan et al 2000] and is being tested for parallel computer performance [Adve et al 2000;Verykios et al 1999].…”

“…To address the complexity issue together with scalability and portability of this approach, we present a knowledge discovery in databases (KDD) methodology [Fayyad et al 1996] for testing and recommending scientific software. PYTHIA-II is a system with an open software architecture implementing the KDD methodology, which can be used to build a Recommender System (RS) for many domains of scientific software/hardware artifacts [Weerawarana et al 1997;Ramakrishnan et al 2000;Verykios 1999;Verykios et al 2000]. In this paper, we describe the PYTHIA-II architecture and its use as an RS for PDE software.…”

Networked agents for scientific computing

“…Each recommender agent can provide recommendations for a certain class of problems and can The PYTHIA system provides the recommender agents needed for multidisciplinary simulation. A PYTHIA agent [12] gathers performance information about solvers on standardized test problems and uses this data (plus features of existing problems) to determine good algorithms to solve a newly presented problem. The efficacy of a single agent is thus dependent on the methods and the problem sets referenced in the performance information collected by it and its ability to determine the features of the new problem.…”

Mining and visualizing recommendation spaces for elliptic PDEs with continuous attributes

“…Rice has been a leader in building frameworks in which large performance-evaluation studies could be conducted and from which new insights into the algorithm selection problem could be gained, e.g., an early PDE solving performance evaluation system [Boisvert et al 1979], the ELLPACK system [Rice and Boisvert 1985], a population of parameterized test problems [Rice et al 1981], and a population of parameterized PDE domains and solutions [Rice 1984;Ribbens and Rice 1986]. More recently, approaches to solving the algorithm selection problem whose roots are in the artificial intelligence community have been developed [Addison et al 1991;Lucks and Gladwell 1992;Kamel et al 1993;Weerawarana et al 1996;Houstis et al 2000]. A particular recent emphasis has been on the implementation of algorithm recommender systems in specialized domains.…”