Livermore Big Artificial Neural Network Toolkit

RESOURCE

Abstract

LBANN is a toolkit that is designed to train artificial neural networks efficiently on high performance computing architectures. It is optimized to take advantages of key High Performance Computing features to accelerate neural network training. Specifically it is optimized for low-latency, high bandwidth interconnects, node-local NVRAM, node-local GPU accelerators, and high bandwidth parallel file systems. It is built on top of the open source Elemental distributed-memory dense and spars-direct linear algebra and optimization library that is released under the BSD license. The algorithms contained within LBANN are drawn from the academic literature and implemented to work within a distributed-memory framework.
Developers:
Essen, Brian [1] Jacobs, Sam [1] Kim, Hyojin [1] Dryden, Nikoli [1] Moon, Tim [1]
  1. Lawrence Livermore National Laboroatory
Release Date:
2016-07-18
Project Type:
Open Source, Publicly Available Repository
Software Type:
Scientific
Licenses:
Apache License 2.0
Sponsoring Org.:
Code ID:
4424
Site Accession Number:
7075
Research Org.:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
Country of Origin:
United States

RESOURCE

Citation Formats

Essen, Brian V., Jacobs, Sam, Kim, Hyojin, Dryden, Nikoli, and Moon, Tim. Livermore Big Artificial Neural Network Toolkit. Computer Software. https://github.com/LLNL/LBANN. USDOE Office of Science (SC). 18 Jul. 2016. Web. doi:10.11578/dc.20171025.1799.
Essen, Brian V., Jacobs, Sam, Kim, Hyojin, Dryden, Nikoli, & Moon, Tim. (2016, July 18). Livermore Big Artificial Neural Network Toolkit. [Computer software]. https://github.com/LLNL/LBANN. https://doi.org/10.11578/dc.20171025.1799.
Essen, Brian V., Jacobs, Sam, Kim, Hyojin, Dryden, Nikoli, and Moon, Tim. "Livermore Big Artificial Neural Network Toolkit." Computer software. July 18, 2016. https://github.com/LLNL/LBANN. https://doi.org/10.11578/dc.20171025.1799.
@misc{ doecode_4424,
title = {Livermore Big Artificial Neural Network Toolkit},
author = {Essen, Brian V. and Jacobs, Sam and Kim, Hyojin and Dryden, Nikoli and Moon, Tim},
abstractNote = {LBANN is a toolkit that is designed to train artificial neural networks efficiently on high performance computing architectures. It is optimized to take advantages of key High Performance Computing features to accelerate neural network training. Specifically it is optimized for low-latency, high bandwidth interconnects, node-local NVRAM, node-local GPU accelerators, and high bandwidth parallel file systems. It is built on top of the open source Elemental distributed-memory dense and spars-direct linear algebra and optimization library that is released under the BSD license. The algorithms contained within LBANN are drawn from the academic literature and implemented to work within a distributed-memory framework.},
doi = {10.11578/dc.20171025.1799},
url = {https://doi.org/10.11578/dc.20171025.1799},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20171025.1799}},
year = {2016},
month = {jul}
}