Search NASA⌕ Search

SEARCH · Search NASA

Results for “code sharing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 361 records · Page 20
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Functional Characterization of Four Putative δ1-Pyrroline-5-Carboxylate Reductases from Bacillus subtilis

In most living organisms, the amino acid proline is synthesized starting from both glutamate and ornithine. In prokaryotes, in the absence of an ornithine cyclodeaminase that has been identified to date only in a small number of soil and plant bacteria, these pathways share the last step, the reduction of δ1-pyrroline-5-carboxylate (P5C) catalyzed by P5C reductase (EC 1.5.1.2). In several species, multiple forms of P5C reductase have been reported, possibly reflecting the dual function of proline. Aside from its common role as a building block of proteins, proline is indeed also involved in the cellular response to osmotic and oxidative stress conditions. Genome analysis of Bacillus subtilis identifies the presence of four genes (ProH, ProI, ProG, and ComER) that, based on bioinformatic and phylogenic studies, were defined as respectively coding a putative P5C reductase. Here we describe the cloning, heterologous expression, functional analysis and small-angle X-ray scattering studies of the four affinity-purified proteins. Results showed that two of them, namely ProI and ComER, lost their catalytic efficiency or underwent subfunctionalization. In the case of ComER, this could be likely explained by the loss of the ability to form a dimer, which has been previously shown to be an essential structural feature of the catalytically active P5C reductase. The properties of the two active enzymes are consistent with a constitutive role for ProG, and suggest that ProH expression may be beneficial to satisfy an increased need for proline.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Open Chemistry, JupyterLab, REST, and quantum chemistry

Quantum chemistry must evolve if it wants to fully leverage the benefits of the internet age, where the worldwide web offers a vast tapestry of tools that enable users to communicate and interact with complex data at the speed and convenience of a button press. The Open Chemistry project has developed an open-source framework that offers an end-to-end solution for producing, sharing, and visualizing quantum chemical data interactively on the web using an array of modern tools and approaches. These tools build on some of the best open-source community projects such as Jupyter for interactive online notebooks, coupled with 3D accelerated visualization, state-of-the-art computational chemistry codes including NWChem and Psi4, and emerging machine learning and data mining tools such as ChemML and ANI. They offer flexible formats to import and export data, along with approaches to compare computational and experimental data.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Counting Kmers for Biological Sequences at Large Scale

Counting the abundance of all the distinct kmers in biological sequence data is a fundamental step in bioinformatics. These applications include de novo genome assembly, error correction, etc. With the development of sequencing technology, the sequence data in a single project can reach Petabyte-scale or Terabyte-scale nucleotides. Counting demand for the abundance of these sequencing data is beyond the memory and computing capacity of single computing node, and how to process it efficiently is a challenge on a high-performance computing cluster. As such, we propose SWAPCounter, a highly scalable distributed approach for kmer counting. This approach is embedded with an MPI streaming I/O module for loading huge data set at high speed, and a counting bloom filter module for both memory and communication efficiency. By overlapping all the counting steps, SWAPCounter achieves high scalability with high parallel efficiency. In this work, the experimental results indicate that SWAPCounter has competitive performance with two other tools on shared memory environment, KMC 2 , and MSPKmerCounter. Moreover, SWAPCounter also shows the highest scalability under strong scaling experiments. In our experiment on Cetus supercomputer, SWAPCounter scales to 32,768 cores with 79% parallel efficiency (using 2048 cores as baseline) when processing 4 TB sequence data of 1000 Genomes. The source code of SWAPCounter is publicly available at https://github.com/mengjintao/SWAPCounter.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2017

Why Vascular Surgeons and Interventional Radiologists Collaborate or Compete: A Look at Endovascular Stent Placements

PurposeTo understand how cultural differences between vascular surgeons (VSs) and interventional radiologists (IRs) affect their clinical decision making and inter-specialty relationships.MethodsTwenty-four conversational interviews were conducted with IRs and VSs about their approaches to patient care, views of their specialty and others, and solutions to any expressed concerns. Interview transcripts were systematically analyzed to identify and compare key themes according to the constructivist grounded theory and content analysis using NVivo 10 software. These data were supplemented with a retrospective analysis of 3658 endovascular stent placements performed at a large medical academic center over 11 years. Aggregate counts were divided by provider specialty, and trends were assessed via correlation coefficients.ResultsEndovascular stent placements were relatively equally divided between IR and VS over 11 years with some variability from placements by cardiology. IRs tend to lay claim to treatments as masters of procedures, whereas VSs base their claims on being masters of the treated diseases, leading to collaboration in some practices and bitter competition in others. The level of perceived competition was most associated with specialists’ awareness of and appreciation for specialty-specific values rather than differences in practice structure/reimbursement.ConclusionsUnderstanding cultural differences between IRs and VSs is imperative for fostering better collaboration to grow shared territory rather than competing for the same slice of the pie.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2022

iMaNGA: mock MaNGA galaxies based on IllustrisTNG and MaStar SSPs – I. Construction and analysis of the mock data cubes

ABSTRACT Galaxy formation and evolution simulations are essential tools to probe poorly known astrophysics processes, but particular care is needed to compare simulations with galaxy observations, as observed data need to be modelled as well. We present a method to generate mock galaxies from the hydro-dynamical IllustrisTNG simulations which are suited to compare with integral field spectroscopic observation of galaxies from the SDSS-IV/Mapping Nearby Galaxies at Apache Point Observatory (MaNGA) survey. First, we include the same instrumental effects and procedures as adopted in the acquisition and analysis of real data. Furthermore, we generate the galaxy spectra from the simulations using new stellar population models based on the MaNGA stellar library (MaStar). In this way, our mock data cubes have the same spatial sampling, cover the same wavelength range (3600–10 300 Å), and share the same spectral resolution (R ≈ 1800) and flux calibration of real MaNGA galaxy spectra. In this first paper, we demonstrate the method over an early- and a late-type simulated galaxy from TNG50. We analyse the correspondent mock MaNGA-like data cubes with the same full spectral fitting code, firefly, which was used for the observed spectra. We find that the intrinsic and recovered age and metallicity gradients are consistent within 1σ, with residuals over all tassels consistent with 0 at the 68 ${{\ \rm per\ cent}}$ confidence level. We also perform the challenging test at comparing intrinsic and recovered star formation histories, finding a close resemblance between input and output. In follow-up papers, we will present a full simulated MaNGA-like catalogue (≈10 000 galaxies) with a comprehensive comparison of TNG50 simulations to MaNGA observational results.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Effect of Lateral Conduction in a Rectangular Fuel Plate and Optimum Stripe Widths for the OPAL Reactor

The impetus for this work is the Proliferation Resistance Optimization (PRO-X) program, which was created by the National Nuclear Security Administration (NNSA) of the US Department of Energy (DOE) to provide a framework for developing reactor designs that minimize the production of special nuclear materials while maximizing performance for peaceful uses. The Argentine company INVAP and Argonne National Laboratory collaborated to share knowledge and expertise regarding the thermal analysis of research reactors. Working together, they designed a series of test problems based on the Open Pool Australian Light Water (OPAL) reactor, whose fuel is in the form of flat plates cooled by water flowing through narrow rectangular channels. ANL used their PLTEMP/ANL code. INVAP used their TERMIC code (and its new multi-plate version, TERMIC-MP). Then, INVAP and ANL compared the two sets of results. 1-D models were used during the collaboration since for research reactor analysis, largely 1-D thermal-hydraulics models are commonly used because they are simple to apply and technically defendable. The collaboration between the two organizations aims for improvements in thermal-hydraulic models. Improvements to methods and models can reduce modeling uncertainties and excessive conservatism, allowing greater reactor performance without reducing predicted safety margins. Therefore, these improvements potentially expand the design space, making additional designs feasible, which, in turn, may achieve better fuel utilization and proliferation resistance in the designs. This report provides justification for the lateral node size (stripe width) chosen for the 1-D computer models during the collaboration. The lateral heat conduction in OPAL fuel plates is evaluated using the two-dimensional conduction capability of PLTEMP/ANL to determine a representative hot-stripe width that can be conservatively used for one-dimensional steady-state thermal-hydraulic analysis. Basically, a representative hot-stripe width needs to be selected so that the actual heat flux averaged over this stripe width is greater than or equal to the maximum heat flux obtained by the two-dimensional analysis. Detailed results in this report show that a hot-stripe width of 8.125 mm can be conservatively used for the one-dimensional thermal-hydraulics analysis of all plates. Therefore, eight equal stripes (8.125 mm each) are recommended to be modeled in the MCNP and PLTEMP/ANL full core models of the OPAL reactor.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2017

Roofline Analysis in the Intel® Advisor to Deliver Optimized Performance for applications on Intel® Xeon Phi™ Processor

In this session we show, in two case studies, how the roofline feature of Intel Advisor has been utilized to optimize the performance of kernels of the XGC1 and PICSAR codes in preparation for Intel Knights Landing architecture. The impact of the implemented optimizations and the benefits of using the automatic roofline feature of Intel Advisor to study performance of large applications will be presented. This demonstrates an effective optimization strategy that has enabled these science applications to achieve up to 4.6 times speed-up and prepare for future exascale architectures. # Goal/Relevance of Session The roofline model [1,2] is a powerful tool for analyzing the performance of applications with respect to the theoretical peak achievable on a given computer architecture. It allows one to graphically represent the performance of an application in terms of operational intensity, i.e. the ratio of flops performed and bytes moved from memory in order to guide optimization efforts. Given the scale and complexity of modern science applications, it can often be a tedious task for the user to perform the analysis on the level of functions or loops to identify where performance gains can be made. With new Intel tools, it is now possible to automate this task, as well as base the estimates of peak performance on measurements rather than vendor specifications. The goal of this session is to demonstrate how the roofline feature of Intel Advisor can be used to balance memory vs. computation related optimization efforts and effectively identify performance bottlenecks. A series of typical optimization techniques: cache blocking, structure refactoring, data alignment, and vectorization illustrated by the kernel cases will be addressed. # Description of the codes ## XGC1 The XGC1 code [3] is a magnetic fusion Particle-In-Cell code that uses an unstructured mesh for its Poisson solver that allows it to accurately resolve the edge plasma of a magnetic fusion device. After recent optimizations to its collision kernel [4], most of the computing time is spent in the electron push (pushe) kernel, where these optimization efforts have been focused. The kernel code scaled well with MPI+OpenMP but had almost no automatic compiler vectorization, in part due to indirect memory addresses and in part due to low trip counts of low-level loops that would be candidates for vectorization. Particle blocking and sorting have been implemented to increase trip counts of low-level loops and improve memory locality, and OpenMP directives have been added to vectorize compute-intensive loops that were identified by Advisor. The optimizations have improved the performance of the pushe kernel 2x on Haswell processors and 1.7x on KNL. The KNL node-for-node performance has been brought to within 30% of a NERSC Cori phase I Haswell node and we expect to bridge this gap by reducing the memory footprint of compute intensive routines to improve cache reuse. ## PICSAR is a Fortran/Python high-performance Particle-In-Cell library targeting at MIC architectures first designed to be coupled with the PIC code WARP for the simulation of laser-matter interaction and particle accelerators. PICSAR also contains a FORTRAN stand-alone kernel for performance studies and benchmarks. A MPI domain decomposition is used between NUMA domains and a tile decomposition (cache-blocking) handled by OpenMP has been added for shared-memory parallelism and better cache management. The so-called current deposition and field gathering steps that compose the PIC time loop constitute major hotspots that have been rewritten to enable more efficient vectorization. Particle communications between tiles and MPI domain has been merged and parallelized. All considered, these improvements provide speedups of 3.1 for order 1 and 4.6 for order 3 interpolation shape factors on KNL configured in SNC4 quadrant flat mode. Performance is similar between a node of cori phase 1 and KNL at order 1 and better on KNL by a factor 1.6 at order 3 with the considered test case (homogeneous thermal plasma).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Draft genome sequence of type strain HBR26T and description of Rhizobium aethiopicum sp. nov.

Rhizobium aethiopicum sp. nov. is a newly proposed species within the genus Rhizobium. This species includes six rhizobial strains; which were isolated from root nodules of the legume plant Phaseolus vulgaris growing in soils of Ethiopia. The species fixes nitrogen effectively in symbiosis with the host plant P. vulgaris, and is composed of aerobic, Gram-negative staining, rod-shaped bacteria. The genome of type strain HBR26 T of R. aethiopicum sp. nov. was one of the rhizobial genomes sequenced as a part of the DOE JGI 2014 Genomic Encyclopedia project designed for soil and plant-associated and newly described type strains. The genome sequence is arranged in 62 scaffolds and consists of 6,557,588 bp length, with a 61% G + C content and 6221 protein-coding and 86 RNAs genes. The genome of HBR26 T contains repABC genes (plasmid replication genes) homologous to the genes found in five differen t Rhizobium etli CFN42 T plasmids, suggesting that HBR26 T may have five additional replicons other than the chromosome. In the genome of HBR26 T , the nodulation genes nodB, nodC, nodS, nodI, nodJ and nodD are located in the same module, and organized in a similar way as nod genes found in the genome of other known common bean-nodulating rhizobial species. nodA gene is found in a different scaffold, but it is also very similar to nodA genes of other bean-nodulating rhizobial strains. Though HBR26 T is distinct on the phylogenetic tree and based on ANI analysis (the highest value 90.2% ANI with CFN42 T ) from other bean-nodulating species, these nod genes and most nitrogen-fixing genes found in the genome of HBR26 T share high identity with the corresponding genes of known bean-nodulating rhizobial species (96-100% identity). This suggests that symbiotic genes might be shared between bean-nodulating rhizobia through horizontal gene transfer. R. aethiopicum sp. nov. was grouped into the genus Rhizobium but was distinct from all recognized species of that genus by phylogenetic analyses of combined sequences of the housekeeping genes recA and glnII. The closest reference type strains for HBR26 T were R. etli CFN42 T (94% similarity of the combined recA and glnII sequences) and Rhizobium bangladeshense BLR175 T (93%). Genomic ANI calculation based on protein-coding genes also revealed that the closest reference strains were R. bangladeshense BLR175 T and R. etli CFN42 T with ANI values 91.8 and 90.2%, respectively. Nevertheless, the ANI values between HBR26 T and BLR175 T or CFN42 T are far lower than the cutoff value of ANI ( > = 96%) between strains in the same species, confirming that HBR26 T belongs to a novel species. Thus, on the basis of phylogenetic, comparative genomic analyses and ANI results, we formally propose the creation of R. aethiopicum sp. nov. with strain HBR26 T (=HAMBI 3550 T =LMG 29711 T ) as the type strain. The genome assembly and annotation data is deposited in the DOE JGI portal and also available at European Nucleotide Archive under accession numbers FMAJ01000001-FMAJ01000062.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Modeling the asynchronous Jacobi method without communication delays

Asynchronous iterative methods for solving linear systems are gaining renewed interest due to the high cost of synchronization points in massively parallel codes. Historically, theory on asynchronous iterative methods has focused on asymptotic behavior, while the transient behavior remains poorly understood. In this paper, we study a model of the asynchronous Jacobi method without communication delays, which we call simplified asynchronous Jacobi. Simplified asynchronous Jacobi can be used to model asynchronous Jacobi implemented in shared memory or distributed memory with fast communication networks. Our analysis uses the idea of a propagation matrix, which is similar in concept to an iteration matrix. We show that simplified asynchronous Jacobi can continue to reduce the residual when some processes are slower than other processes. We also show that simplified asynchronous Jacobi can converge when synchronous Jacobi does not. We verify our analysis of simplified asynchronous Jacobi using results from asynchronous Jacobi implemented in shared and distributed memory.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Pseudouridine synthase 7 is an opportunistic enzyme that binds and modifies substrates with diverse sequences and structures

Pseudouridine (Ψ) is a ubiquitous RNA modification incorporated by pseudouridine synthase (Pus) enzymes into hundreds of noncoding and protein-coding RNA substrates. Here, in this study, we determined the contributions of substrate structure and protein sequence to binding and catalysis by pseudouridine synthase 7 (Pus7), one of the principal messenger RNA (mRNA) modifying enzymes. Pus7 is distinct among the eukaryotic Pus proteins because it modifies a wider variety of substrates and shares limited homology with other Pus family members. We solved the crystal structure of Saccharomyces cerevisiae Pus7, detailing the architecture of the eukaryotic-specific insertions thought to be responsible for the expanded substrate scope of Pus7. Additionally, we identified an insertion domain in the protein that fine-tunes Pus7 activity both in vitro and in cells. These data demonstrate that Pus7 preferentially binds substrates possessing the previously identified UGUAR (R = purine) consensus sequence and that RNA secondary structure is not a strong requirement for Pus7-binding. In contrast, the rate constants and extent of Ψ incorporation are more influenced by RNA structure, with Pus7 modifying UGUAR sequences in less-structured contexts more efficiently both in vitro and in cells. Although less-structured substrates were preferred, Pus7 fully modified every transfer RNA, mRNA, and nonnatural RNA containing the consensus recognition sequence that we tested. Our findings suggest that Pus7 is a promiscuous enzyme and lead us to propose that factors beyond inherent enzyme properties (e.g., enzyme localization, RNA structure, and competition with other RNA-binding proteins) largely dictate Pus7 substrate selection.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Snippet Open-Source Specification Synopsis

Snippet is a collaborative source code auditing tool developed by Sandia National Laboratories, in conjunction with researchers from other government agencies and partners. As a tool for use by groups of auditors, Snippet showed the value of providing in-line annotations that captured the thoughts of auditors in such a way that teams of auditors could share their insights. Snippet users found that the annotation mechanisms improved the auditing process but were frustrated at having to learn another environment. To address this issue, collaborators made extensions to existing Integrated Development Environments (IDEs) to incorporate functionality that mimics Snippet's behaviors. This document is one of a pair of documents. The other document, "Snippet Open-Source Specification", goes into details about the history and reasoning behind the requirements made in this document. That document provides for a broader understanding of the motivations backing this document. This document provides the complete specification for collaborative annotation, and a characterization of each specification as required or recommended. Tools that implement these specifications will be considered Snippet compliant. The intent is that complete implementation of these specifications will enable seamless sharing of annotation information between disjoint tools.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Developing and Distributing HEP Software Stacks with Spack

The Computational Science and AI Directorate at Fermilab is using Spack to support the development efforts of a large number of scientific programmers, in many independent projects and experiments. While independent, these projects share many dependencies. They are typically under continuous and fairly rapid development. They have to support deployment on diverse hardware. This is a different context than is typical for the management of HPC software, where Spack was born. To support our community, we have created a model that enables users to develop code with greater efficiency than is possible with Spack’s current development facilities. In this talk we will present: - a brief introduction to the science we support (particle physics) - how the code we work with is naturally organized into several layers of packages - how we are using Spack to manage those layers - how we leverage the layering to provide efficient support for developers, using our Spack extension “MPD”. - some suggestions for changes or additions to Spack to make such work easier.

Knoepfel, Kyle J. [Fermilab]↗
DOE OSTI2015

Genetic algorithm based task reordering to improve the performance of batch scheduled massively parallel scientific applications

The growth in size of networked high performance computers along with novel accelerator-based node architectures has further emphasized the importance of communication efficiency in high performance computing. The world's largest high performance computers are usually operated as shared user facilities due to the costs of acquisition and operation. Applications are scheduled for execution in a shared environment and are placed on nodes that are not necessarily contiguous on the interconnect. Furthermore, the placement of tasks on the nodes allocated by the scheduler is sub-optimal, leading to performance loss and variability. Here, we investigate the impact of task placement on the performance of two massively parallel application codes on the Titan supercomputer, a turbulent combustion flow solver (S3D) and a molecular dynamics code (LAMMPS). Benchmark studies show a significant deviation from ideal weak scaling and variability in performance. The inter-task communication distance was determined to be one of the significant contributors to the performance degradation and variability. A genetic algorithm-based parallel optimization technique was used to optimize the task ordering. This technique provides an improved placement of the tasks on the nodes, taking into account the application's communication topology and the system interconnect topology. As a result, application benchmarks after task reordering through genetic algorithm show a significant improvement in performance and reduction in variability, therefore enabling the applications to achieve better time to solution and scalability on Titan during production.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

The Earth System Prediction Suite: Toward a Coordinated U.S. Modeling Capability

The Earth System Prediction Suite (ESPS) is a collection of flagship U.S. weather and climate models and model components that are being instrumented to conform to interoperability conventions, documented to follow metadata standards, and made available either under open-source terms or to credentialed users. Furthermore, the ESPS represents a culmination of efforts to create a common Earth system model architecture, and the advent of increasingly coordinated model development activities in the United States. ESPS component interfaces are based on the Earth System Modeling Framework (ESMF), community-developed software for building and coupling models, and the National Unified Operational Prediction Capability (NUOPC) Layer, a set of ESMF-based component templates and interoperability conventions. Our shared infrastructure simplifies the process of model coupling by guaranteeing that components conform to a set of technical and semantic behaviors. The ESPS encourages distributed, multiagency development of coupled modeling systems; controlled experimentation and testing; and exploration of novel model configurations, such as those motivated by research involving managed and interactive ensembles. ESPS codes include the Navy Global Environmental Model (NAVGEM), the Hybrid Coordinate Ocean Model (HYCOM), and the Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS); the NOAA Environmental Modeling System (NEMS) and the Modular Ocean Model (MOM); the Community Earth System Model (CESM); and the NASA ModelE climate model and the Goddard Earth Observing System Model, version 5 (GEOS-5), atmospheric general circulation model.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Development of MCNP Training Modules for Safeguards Practitioners [Abstract]

The Monte-Carlo N-Particle (MCNP) software developed at LANL is the most widely used neutron transport code in the world. It is an essential tool for a variety of applications including detector development and design, nuclear fuel burnup simulation, criticality safety, and nondestructive assay system optimization. For this reason, it is indispensable within the safeguards and materials control & accountability (MC&A) communities. Multiple MCNP training courses have been created and taught over the last several decades by the MCNP development team at LANL, however there are no existing courses that cover specialized topics considered fundamental to NDA and safeguards models. To fill this gap, the MCNP team and Safeguards Science and Technology group at LANL have co-created a set of training modules customized to meet the specialized needs of the safeguards and MC&A communities. The basic modules cover concepts such as NDA system optimization, He-specific and other capture tallies, and tools for improved theoretical understanding. An advanced module was also created to cover topics including variance reduction for active interrogation simulations, use of the LANL MCNPTools post-processor, PTRAC (particle tracking) and list-mode data simulations, and fuel burnup simulations. The training modules teach to the latest and most state-of-the-art MCNP features and tools released by the development team at LANL and are intended to be taught jointly by the developers and safeguards experts. Ultimately, we hope that creation of these modules will serve to capture and convey the safeguards modeling and MCNP expertise at LANL, and that we will be able to share the modules more broadly with the MC&A and safeguards communities.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2019

Computer Modeling and Simulation: Increasing Reliability by Disentangling Verification and Validation

Verification and validation (V&V) of computer codes and models used in simulations are two aspects of the scientific practice of high importance that recently have been discussed widely by philosophers of science. While verification is predominantly associated with the correctness of the way a model is represented by a computer code or algorithm, validation more often refers to the model’s relation to the real world and its intended use. Because complex simulations are generally opaque to a practitioner, the Duhem problem can arise with verification and validation due to their entanglement; such an entanglement makes it impossible to distinguish whether a coding error or the model’s general inadequacy to its target should be blamed in the case of a failure. I argue that a clear distinction between computer modeling and simulation has to be made to disentangle verification and validation. Drawing on that distinction, I suggest to associate modeling with verification and simulation, which shares common epistemic strategies with experimentation, with validation. To explain the reasons for their entanglement in practice, I propose a Weberian ideal–typical model of modeling and simulation as roles in practice. I examine an approach to mitigate the Duhem problem for verification and validation that is generally applicable in practice and is based on differences in epistemic strategies and scopes. Based on this analysis, I suggest two strategies to increase the reliability of simulation results, namely, avoiding alterations of verified models at the validation stage as well as performing simulations of the same target system using two or more different models. In response to Winsberg’s claim that verification and validation are entangled I argue that deploying the methodology proposed in this work it is possible to mitigate inseparability of V&V in many if not all domains where modeling and simulation are used.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

A generalized massively parallel ultra-high order FFT-based Maxwell solver

Dispersion-free ultra-high order FFT-based Maxwell solvers have recently proven to be paramount to a large range of applications, including the high-fidelity modeling of high-intensity laser–matter interactions with Particle-In-Cell (PIC) codes. To enable a massively parallel scaling of these solvers, a novel parallelization technique was recently proposed, which consists in splitting the simulation domain into several processor sub-domains, with guard regions appended at each sub-domain boundary. Maxwell's equations are advanced independently on each sub-domain using local shared-memory FFTs (instead of a single distributed global FFT). This implies small truncation errors at sub-domain boundaries, the amplitude of which depends on guard regions sizes and order of the Maxwell solver. For moderate guard region sizes, this ’local’ technique proved to be highly scalable on up to a million cores and notably enabled the 3D modeling of so-called plasma mirrors, for which 8 guard cells only were enough to prevent truncation error growth. Yet, for other applications, the required number of guard cells might be much higher, which would severely limit the parallel efficiency of this technique due to the large volume of guard cells to be exchanged between sub-domains. In this context, we propose a novel parallelization technique that ensures very good scaling of FFT-based solvers with an arbitrarily high number of guard cells. Our ’hybrid’ technique consists in performing distributed FFTs on local groups of processors with guard regions now appended to boundaries of each group of processors. It uses a dual domain decomposition method for the Maxwell solver and other parts of the PIC cycle to keep the simulation load-balanced. This ’hybrid’ technique was implemented in the open source exascale library PICSAR. Benchmarks show that for a large number of guard cells (>16), the ’hybrid’ technique offers up to ×3 speed-up and ×8 memory savings compared to the ’local’ one.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2019

Sierra Center of Excellence: Lessons Learned

The introduction of heterogeneous computing via GPUs from the Sierra architecture represented a significant shift in direction for computational science at Lawrence Livermore National Laboratory (LLNL), and therefore required significant preparation. Over the last five years, the Sierra Center of Excellence (CoE) has brought employees with specific expertise from IBM and NVIDIA together with LLNL in a concentrated effort to prepare applications, system software, and tools for the Sierra supercomputer. This article shares the process we applied for the CoE and documents lessons learned during the collaboration, with the hope that others will be able to learn from both our success and intermediate setbacks. We describe what we have found to work for the management of such a collaboration and best practices for algorithms and source code, system configuration and software stack, tools, and application performance.

97 MATHEMATICS AND COMPUTING↗