Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems

Thomas, Mathew; Schram, Malachi; Fox, Kevin; Strube, Jan; Oblath, Noah S.; Rallo, Robert; Kennedy, Zachary C.; Varga, Tamas; Battu, Anil K.; Barrett, Christopher A.

doi:10.1557/adv.2020.103

Title: Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems

Abstract

In this work, we present the current status of a large-scale computing framework to address the need of the multidisciplinary effort to study chemical dynamics. Specifically, we are enabling scientists to process and store experimental data, run large-scale computationally expensive high-fidelity physical simulations, and analyze these results using the state-of-the-art data analytics tools, machine learning, and uncertainty quantification methods using heterogeneous computing resources, such as CPU and GPU cluster. The framework can integrate or abstract out multiple domains based on roles. In order to develop this framework, we have leveraged an existing framework coupled with in-house heterogeneous computing resources. We present the results of using this framework on a single metadata triggered workflow to accelerate an additive manufacturing use case.

Authors:

Thomas, Mathew ^[1]; Schram, Malachi ^[1]; Fox, Kevin ^[1]; Strube, Jan ^[1]; Oblath, Noah S. ^[1]; Rallo, Robert ^[1]; Kennedy, Zachary C. ^[1]; Varga, Tamas ^[1]; Battu, Anil K. ^[1]; Barrett, Christopher A. ^[1]

Pacific Northwest National Lab. (PNNL), Richland, WA (United States)

Publication Date:: 2020-02-07

Research Org.:: Pacific Northwest National Lab. (PNNL), Richland, WA (United States)

Sponsoring Org.:: USDOE

OSTI Identifier:: 1605909

Alternate Identifier(s):: OSTI ID: 1657189

Report Number(s):: PNNL-SA-150139
Journal ID: ISSN 2059-8521

Grant/Contract Number:: AC05-76RL01830

Resource Type:: Accepted Manuscript

Journal Name:: MRS Advances

Additional Journal Information:: Journal Volume: 5; Journal Issue: 29-30; Journal ID: ISSN 2059-8521

Publisher:: Materials Research Society (MRS)

Country of Publication:: United States

Language:: English

Subject:: 36 MATERIALS SCIENCE; additive manufacturing; infrastructure; data

Citation Formats


                    Thomas, Mathew, Schram, Malachi, Fox, Kevin, Strube, Jan, Oblath, Noah S., Rallo, Robert, Kennedy, Zachary C., Varga, Tamas, Battu, Anil K., and Barrett, Christopher A. Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems.  United States: N. p., 2020. 
Web.  doi:10.1557/adv.2020.103.

Copy to clipboard


                    Thomas, Mathew, Schram, Malachi, Fox, Kevin, Strube, Jan, Oblath, Noah S., Rallo, Robert, Kennedy, Zachary C., Varga, Tamas, Battu, Anil K., & Barrett, Christopher A. Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems.  United States.  https://doi.org/10.1557/adv.2020.103

Copy to clipboard


                    Thomas, Mathew, Schram, Malachi, Fox, Kevin, Strube, Jan, Oblath, Noah S., Rallo, Robert, Kennedy, Zachary C., Varga, Tamas, Battu, Anil K., and Barrett, Christopher A. Fri .  
"Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems".  United States.  https://doi.org/10.1557/adv.2020.103.  https://www.osti.gov/servlets/purl/1605909.

Copy to clipboard


                    
@article{osti_1605909,

  title        = {Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems},

  author       = {Thomas, Mathew and Schram, Malachi and Fox, Kevin and Strube, Jan and Oblath, Noah S. and Rallo, Robert and Kennedy, Zachary C. and Varga, Tamas and Battu, Anil K. and Barrett, Christopher A.},

  abstractNote = {In this work, we present the current status of a large-scale computing framework to address the need of the multidisciplinary effort to study chemical dynamics. Specifically, we are enabling scientists to process and store experimental data, run large-scale computationally expensive high-fidelity physical simulations, and analyze these results using the state-of-the-art data analytics tools, machine learning, and uncertainty quantification methods using heterogeneous computing resources, such as CPU and GPU cluster. The framework can integrate or abstract out multiple domains based on roles. In order to develop this framework, we have leveraged an existing framework coupled with in-house heterogeneous computing resources. We present the results of using this framework on a single metadata triggered workflow to accelerate an additive manufacturing use case.},

  doi          = {10.1557/adv.2020.103},

  journal      = {MRS Advances},

  number       = 29-30,

  volume       = 5,

  place        = {United States},

  year         = {2020},

  month        = {2}

}

Copy to clipboard

Journal Article:

Free Publicly Available Full Text

Accepted Manuscript (DOE)

Publisher's Version of Record

https://doi.org/10.1557/adv.2020.103

Other availability

Search WorldCat to find libraries that may hold this journal

Save / Share:

Export Metadata

Save to My Library

Works referenced in this record:

Fiji: an open-source platform for biological-image analysis
journal, June 2012

Schindelin, Johannes; Arganda-Carreras, Ignacio; Frise, Erwin
Nature Methods, Vol. 9, Issue 7
DOI: 10.1038/nmeth.2019

Similar Records in DOE PAGES and OSTI.GOV collections:

Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems

Journal Article Thomas, Mathew ; Schram, Malachi ; Fox, Kevin M. ; ... - MRS Advances

We present the current status of a large-scale computing framework to address the need of the multidisciplinary effort to study chemical dynamics. Specifically, we are enabling scientists to process and store experimental data, run large-scale computationally expensive high-fidelity physical simulation, and analyze these results using the state-of-the-art data analytics tools, machine learning, and uncertainty quantification methods using heterogeneous computing resources, such as CPU and GPU cluster. The framework can integrate or abstract out multiple domains based on roles. In order to develop this framework, we have leveraged an existing framework coupled with in-house heterogeneous computing resources. We present the resultsmore »« less
https://doi.org/10.1557/adv.2020.103
Performance modeling of microsecond scale biological molecular dynamics simulations on heterogeneous architectures

Journal Article Agarwal, Pratul K. ; Ramanathan, Arvind - Concurrency and Computation: Practice and Experience

Performance improvements in biomolecular simulations based on molecular dynamics (MD) codes are widely desired. Unfortunately, the factors, which allowed past performance improvements, particularly the microprocessor clock frequencies, are no longer increasing. Hence, novel software and hardware solutions are being explored for accelerating performance of widely used MD codes. In this paper, we describe our efforts on porting, optimizing and tuning of Large-scale Atomic/Molecular Massively Parallel Simulator, a popular MD framework, on heterogeneous architectures: multi-core processors with graphical processing unit (GPU) accelerators. Our implementation is based on accelerating the most computationally expensive non-bonded interaction terms on the GPUs and overlapping themore »« less
https://doi.org/10.1002/cpe.2943
Performance modeling of microsecond scale biological molecular dynamics simulations on heterogeneous architectures.

Journal Article Agarwal, Pratul K. ; Hampton, Scott ; Poznanovic, Jeffrey ; ... - Concurrency and Computation. Practice and Experience

Performance improvements in biomolecular simulations based on molecular dynamics (MD) codes are widely desired. Unfortunately, the factors, which allowed past performance improvements, particularly the microprocessor clock frequencies, are no longer increasing. Hence, novel software and hardware solutions are being explored for accelerating performance of widely used MD codes. In this paper, we describe our efforts on porting, optimizing and tuning of Large-scale Atomic/Molecular Massively Parallel Simulator, a popular MD framework, on heterogeneous architectures: multi-core processors with graphical processing unit (GPU) accelerators. Our implementation is based on accelerating the most computationally expensive non-bonded interaction terms on the GPUs and overlapping themore »« less
https://doi.org/10.1002/cpe.2943
Radiative Heat Transfer Calculation on 16384 GPUs Using a Reverse Monte Carlo Ray Tracing Approach with Adaptive Mesh Refinement

Conference Humphrey, Alan ; Sunderland, Daniel ; Harman, Todd ; ...

Modeling thermal radiation is computationally challenging in parallel due to its all-to-all physical and resulting computational connectivity, and is also the dominant mode of heat transfer in practical applications such as next-generation clean coal boilers, being modeled by the Uintah framework. However, a direct all-to-all treatment of radiation is prohibitively expensive on large computers systems whether homogeneous or heterogeneous. DOE Titan and the planned DOE Summit and Sierra machines are examples of current and emerging GPU-based heterogeneous systems where the increased processing capability of GPUs over CPUs exacerbates this problem. These systems require that computational frameworks like Uintah leverage anmore »« less
https://doi.org/10.1109/IPDPSW.2016.93
A Quantitative Study of Deep Learning Training on Heterogeneous Supercomputers

Conference Han, Jingoo ; Xu, Luna ; Rafique, Mustafa ; ...

Deep learning (DL) has become a key technique for solving complex problems in scientific research and discovery. DL training for science is substantially challenging because it has to deal with massive quantities of multi-dimensional data. High- performance computing (HPC) supercomputers are increasingly being employed for meeting the exponentially growing demand for DL. Multiple GPUs and high-speed interconnect network are needed for supporting DL on HPC systems. However, the excessive use of GPUs without considering effective benefits leads to inefficient resource utilization of these expensive setups. In this paper, we conduct a quantitative analysis to gauge the efficacy of DL workloadsmore »« less
Full Text Available

Similar Records

Title: Distributed heterogeneous compute infrastructure for the study of additive manufacturing systems

Abstract

Citation Formats

Fiji: an open-source platform for biological-image analysis journal, June 2012

Fiji: an open-source platform for biological-image analysis
journal, June 2012