One of the main obstacles to the efficient solution of scientific problems is the problem of tuning software, both to the available architecture and to the user problem at hand. We describe approaches for obtaining tuned high-performance kernels, and for automatically choosing suitable algorithms. Specifically, we describe the generation of dense and sparse blas kernels, and the selection of linear solver algorithms. However, the ideas presented here extend beyond these areas, which can be considered proof of concept.

No abstract

No abstract

We propose a standard for generating, manipulating, and storing metadata describing numerical problems, in particular properties of matrices and linear systems. The standard comprises:-an API for metadata generating and querying software, and -an XML format for permanent storage of metadata.The API is open-ended, allowing for other parties to define additional metadata categories to be generated and stored within this framework. Furthermore, we present two software libraries, NMD and AnaMod, that implement this standard, and that contain a number of computational modules for numerical metadata. The libraries, more than simply illustrating the use of the standard, provide considerable utility to numerical researchers. ACM Reference Format:Eijkhout, V. and Fuentes, E. 2009. A standard and software for numerical metadata. ACM

In various areas of numerical analysis, there are several possible algorithms for solving a problem. In such cases, each method potentially solves the problem, but the runtimes can widely differ, and breakdown is possible. Also, there is typically no governing theory for finding the best method, or the theory is in essence uncomputable. Thus, the choice of the optimal method is in practice determined by experimentation and 'numerical folklore'. However, a more systematic approach is needed, for instance since such choices may need to be made in a dynamic context such as a time-evolving system. Thus we formulate this as a classification problem: assign each numerical problem to a class corresponding to the best method for solving that problem. What makes this an interesting problem for Machine Learning, is the large number of classes, and their relationships. A method is a combination of (at least) a preconditioner and an iterative scheme, making the total number of methods the product of these individual cardinalities. Since this can be a very large number, we want to exploit this structure of the set of classes, and find a way to classify the components of a method separately. We have developed various techniques for such multi-stage recommendations, using automatic recognition of super-clases. These techniques are shown to pay off very well in our application area of iterative linear system solvers. We present the basic concepts of our recommendation strategy, and give an overview of the software libraries that make up the Salsa (Self-Adapting Large-scale Solver Architecture) project.