Approximate Ad-hoc Query Engine for Simulation Data

Ghaleb Abdulla, Chuck Baldwin, Terence Critchlow, Roy Kamimura, Ida Lozares, NuAi Tang
Center for Applied Scientific Computing
Lawrence Livermore National Laboratory
P.O. Box 808, L-561, Livermore, CA 94551
{abdulla1, baldwin5, critchlow, kamimura1, ilozares, tangn}@llnl.gov

Ron Musick*
iKuni, Inc
Palo Alto, CA 94304
musick@ikuni.com

Abstract

In this paper we descibe AQSim, an ongoing effort to design and implement a system to manage terabytes of scientific simulation data. The project is aimed to reduce data storage requirements and access time, and facilitates ad-hoc queries over the data sets. Our approach is to build approximate statistical and mathematical models of the data. In addition to the benefit of data reduction, the models will be used in conjunction with model-related metadata to allow for ad-hoc queries. In order to facilitate data eschange between models based on different representations, we are evaluating layers of increasing semantic complexity. To support queries over the spatial-temporal mesh structured data we are in the process of defining and implementing a grammar for MeshSQL.

To Appear in

The First ACM+IEEE Joint Conference on Digital Libraries, Roanoke Virginia, June 2001.

* Work done while author at
Center for Applied Scientific Computing
Lawrence Livermore National Laboratory
P.O. Box 808, L-561, Livermore, CA 94551