What is Pythia-II? The Pythia-II system is described in PYTHIA II: A Knowledge/Database System for Managing Performance Data and Recommending Scientific Software, to appear in ACM Transactions on Mathematical Software, 2000. Why Pythia-II? Scientists often face the task of locating solution software for their problems, and then selecting from among many alternatives. Pythia-II proposes an approach for dealing with this task by processing performance data from the targeted software. The high level of complexity involved in the algorithmic discovery of knowledge from performance data and in the management of the performance data and discovered knowledge is handled by combining knowledge discovery in databases (KDD) methodologies, and recommender system technologies. How is Pythia-II used? From the end-user perspective, PYTHIA-II allows users to specify the problem to be solved along with their computational objectives, and then selects the software resources available for the user's problem,suggests parameter values, and provides phenomenological assessment of the recommendation provided. For the development of testing software repositories, PYTHIA-II provides all the necessary facilities to set up database schemas for testing suites and associated performance data. It also allows the easy interfacing of alternative data mining and recommendation facilities. PYTHIA-II is an open-ended system implemented on public domain software and can be easily applied to any software domain.
	The Pythia-II User Interface dbEdit, dataGen, dataMine: A knowledge engineer creates the performance database and uses integrated KDD methods for knowledge discovery. Recommender: An end user requests advice about scientific software, and is provided with software/harware selections and parameter information.

Elias Houstis and Ann Christine Catlin, Computer Sciences Department Purdue University West Lafayette, IN. 47907 Partially supported by the National Science Foundation, DARPA, DOE, Intel Corporation and the Purdue Research Foundation.