Computational Science Technical Note CSTN-197

CSTN Home

Performance of Mining Medium-to-Large-Scale Scientific Simulation Data

S. G. Morgan and K. A. Hawick

Archived: 2013

Abstract

Many scientific simulations generate bulky data sets that must be mined for observable features. It is often not computationally feasible to do this in real time and the data must be saved for subsequent ``off line'' analysis either by separate software or sometimes by direct human visualisation. We present some scoping analysis and preliminary software approaches for mining medium to large scale data sets in the form of time slices or model configurations. We report on current storage and visualisation technology response and interaction times for mining scientific simulations on regular lattices using hyper-bricks of model configurations.

Keywords: big data; data mining; data visualization; scientific simulations

Full Document Text: PDF version.

Citation Information: BiBTeX database for CSTN Notes.

BiBTeX reference:

@INPROCEEDINGS{CSTN-197,
        author = {S. G. Morgan and K. A. Hawick},
        title = {Performance of Mining Medium-to-Large-Scale Scientific Simulation
                Data},
        booktitle = {Proc. 14th International Conference on Internet Computing and Big
                Data (ICOMP'13)},
        year = {2013},
        number = {CSTN-197},
        pages = {ICM4063},
        address = {Las Vegas, USA},
        month = {22-25 July},
        publisher = {WorldComp},
        institution = {Computer Science, Massey University, Auckland, New Zealand},
        keywords = {big data; data mining; data visualization; scientific simulations},
        owner = {kahawick},
        timestamp = {2013.04.22}
}


[ CSTN Index | CSTN BiBTeX ]