TuckerCompressMPI v. 1.0

RESOURCE

Abstract

As parallel computing trends towards the exascale, scientific data produced by high-fidelity simulations are growing increasingly massive. For instance, a simulation on a three-dimensional spatial grid with 512 points per dimension that tracks 64 variables per grid point for 128 time steps yields 8 TB of data. By viewing the data as a dense five-way tensor, we can compute a Tucker decomposition to find inherent low-dimensional multilinear structure, achieving compression ratios of up to 10000 on real-world data sets with negligible loss in accuracy. So that we can operate on such massive data, we present the first-ever distributed-memory parallel implementation for the Tucker decomposition, whose key computations correspond to parallel linear algebra operations, albeit with nonstandard data layouts. Our approach specifies a data distribution for tensors that avoids any tensor data redistribution, either locally or in parallel. This software provides a method for compressing large-scale multiway data.
Developers:
Austin, Woody [1] Klinvex, Alicia [1] Ballard, Grey [2] Kolda, Tamara [1]
  1. Sandia National Laboratories
  2. Wake Forest University
Release Date:
2016-09-21
Project Type:
Open Source, Publicly Available Repository
Software Type:
Scientific
Programming Languages:
C++
Shell
CMake
MATLAB
Makefile
Python
Version:
1.0
Licenses:
BSD 2-clause "Simplified" License
Sponsoring Org.:
Code ID:
45231
Site Accession Number:
SCR #2148.0; 7258
Research Org.:
Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)
Country of Origin:
United States

RESOURCE

Citation Formats

Austin, Woody, Klinvex, Alicia, Ballard, Grey, and Kolda, Tamara G. TuckerCompressMPI v. 1.0. Computer Software. https://github.com/sandialabs/TuckerMPI. USDOE. 21 Sep. 2016. Web. doi:10.11578/dc.20201001.6.
Austin, Woody, Klinvex, Alicia, Ballard, Grey, & Kolda, Tamara G. (2016, September 21). TuckerCompressMPI v. 1.0. [Computer software]. https://github.com/sandialabs/TuckerMPI. https://doi.org/10.11578/dc.20201001.6.
Austin, Woody, Klinvex, Alicia, Ballard, Grey, and Kolda, Tamara G. "TuckerCompressMPI v. 1.0." Computer software. September 21, 2016. https://github.com/sandialabs/TuckerMPI. https://doi.org/10.11578/dc.20201001.6.
@misc{ doecode_45231,
title = {TuckerCompressMPI v. 1.0},
author = {Austin, Woody and Klinvex, Alicia and Ballard, Grey and Kolda, Tamara G.},
abstractNote = {As parallel computing trends towards the exascale, scientific data produced by high-fidelity simulations are growing increasingly massive. For instance, a simulation on a three-dimensional spatial grid with 512 points per dimension that tracks 64 variables per grid point for 128 time steps yields 8 TB of data. By viewing the data as a dense five-way tensor, we can compute a Tucker decomposition to find inherent low-dimensional multilinear structure, achieving compression ratios of up to 10000 on real-world data sets with negligible loss in accuracy. So that we can operate on such massive data, we present the first-ever distributed-memory parallel implementation for the Tucker decomposition, whose key computations correspond to parallel linear algebra operations, albeit with nonstandard data layouts. Our approach specifies a data distribution for tensors that avoids any tensor data redistribution, either locally or in parallel. This software provides a method for compressing large-scale multiway data.},
doi = {10.11578/dc.20201001.6},
url = {https://doi.org/10.11578/dc.20201001.6},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20201001.6}},
year = {2016},
month = {sep}
}