A multiresolution data model for improving simulation I/O performance

Author

Foulks, Andrew ; Bergeron, R. Daniel

Author_Institution

Comput. Sci. Dept., Univ. of New Hampshire, Durham, NH, USA

fYear

2011

fDate

18-21 Dec. 2011

Firstpage

1

Lastpage

10

Abstract

Numerical simulations running on very large High Performance Computer clusters still suffer from the I/O bottleneck. The cost of communication can overwhelm the cost of computation, and scales inversely with the number of processors used in the cluster. In previous work we have developed a multiresolution data model to help improve performance for visualizations of very large multi dimensional scientific data sets. In our approach, the data is represented as a multi level hierarchy. Reconstructive error analysis is used to identify regions in the data where the data loss is greatest. We have incorporated this data model into the OpenGGCM solar wind simulation environment. In this paper, we demonstrate that this approach can reduce the I/O and improve the overall performance of a large numerical simulation environment.

Keywords

data models; data visualisation; error analysis; parallel processing; pattern clustering; solar wind; I/O bottleneck; I/O reduction; OpenGGCM solar wind simulation environment; communication cost; data loss; data representation; multidimensional scientific data visualization; multilevel hierarchy; multiresolution data model; numerical simulation environment; reconstructive error analysis; simulation I/O performance; very large high performance computer cluster; Computational modeling; Data models; Data visualization; Numerical models; Rendering (computer graphics); Spatial resolution;

fLanguage

English

Publisher

ieee

Conference_Titel

High Performance Computing (HiPC), 2011 18th International Conference on

Conference_Location

Bangalore

Print_ISBN

978-1-4577-1951-6

Electronic_ISBN

978-1-4577-1949-3

Type

conf

DOI

10.1109/HiPC.2011.6152747

Filename

6152747