SEARCH · Search NASA
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.
High performance computers and the people who use them constitute a distributed cognitive system, with part of the work being done in the machine and part in the brains of developer scientists. The UNH portion of this proposal was devoted to the perceptual and cognitive optimization of workflows and visualizations use in High Performance Computing.
The solidification microstructure in IN718 during additive manufacturing was modeled using phase field simulations. The novelty of the research includes the use of a surrogate Ni–Fe–Nb alloy that has the same equilibrium solidification range as IN718 as the model system for phase field simulations, the integration of the model alloy thermodynamics with the phase field simulations, and the use of high-performance computing tools to perform the simulations with a high enough spatial resolution for realistically capturing the dendrite morphology and the level of microsegregation seen under additive manufacturing conditions. Heat transfer and fluid flow models were used to compute the steady state temperature gradient and an average value of the solid-liquid (s-l) interface velocity that were used as input for the phase field simulations. The simulations show that the solidification morphology is sensitive to the spacing between the columnar structures. Spacing narrower than a critical value results in continued growth of a columnar microstructure, while above a critical value the columnar structure evolves into a columnar dendritic structure through the formation of secondary arms. These results are discussed in terms of the existing columnar to dendritic transition (CDT) theories. The measured interdendritic Nb concentration, the primary and secondary arm spacing is in reasonable agreement with experimental measurements performed on the nickel-base superalloy IN718.
High performance computing (HPC) scheduling landscape currently faces new challenges due to thechanges in the workload. Previously, HPC centers were dominated by tightly coupled MPI jobs. HPCworkloads increasingly include high-throughput, data-intensive, and stream-processing applications. Asa consequence, workloads are becoming more diverse at both application and job levels, posing newchallenges to classical HPC schedulers. There is a need to understand the current HPC workloads andtheir evolution to facilitate informed future scheduling research and enable efficient scheduling in futureHPC systems.In this paper, we present a methodology to characterize workloads and assess their heterogeneity,at a particular time period and its evolution over time. We apply this methodology to the workloads ofthree systems (Hopper, Edison, and Carver) at the National Energy Research Scientific Computing Center(NERSC). We present the resulting characterization of jobs, queues, heterogeneity, and performance thatincludes detailed information of a year of workload (2014) and evolution through the systems’ lifetime(2010–2014).
In the studies of materials, experimental measurements often serve as the reference to verify physics theory and modeling; while theory and modeling provide a fundamental understanding of the physics and principles behind. However, the interactions and cross validation between them have long been a challenge even to-date. Not only that inferring a physics model from experimental data is itself a difficult inverse problem, another major challenge is the orders-of-magnitude longer wall-clock time required to carry out high-fidelity computer modeling to match the timescale of experiments. We envisage that by combining high performance computing, data science, and edge computing technology, the current predicament can be alleviated, and a new paradigm of data-driven physics research will open up. For example, we can accelerate computer simulations by first performing the large-scale modeling on high performance computers and train a machine-learned surrogate model. This computationally inexpensive surrogate model can then be transferred to the computing units residing closely to the experimental facilities to perform high-fidelity simulations at a much higher throughout. The model will also be more amenable to analyzing and validating experimental observations in comparable time scales at a much lower computational cost. Further integration of these accelerated computer simulations with an outer machine learning loop can also inform and direct future experiments, while making the inverse problem of physics model inference more tractable. We will demonstrate a proof-of-concept by using a quantum Monte Carlo application, Dynamical Cluster Approximation (DCA++), to machine-learn a surrogate model and accelerate the study of quantum correlated materials.
Scientific workflows are increasingly relying on machine learning (ML), simulation, and hybrid techniques to predict, understand, and optimize the behavior of complex experiments. High-performance computing has greatly improved researchers’ ability to acquire diverse data modalities in these workflows. Recent studies suggest that the performance of machine learning models can be improved by integrating data from various sources. Unfortunately, these workloads pose unprecedent pressure on the network storage to meet the demands associated with accessing these multimodal data. To mitigate the impact of intensive IO, we propose a solution that utilizes a multi-tier High-Performance Computing (HPC) distributed storage and data processing framework, placing computation where the data resides for better performance. By adopting this project, the scientific community will gain new opportunities to explore multimodal storage-driven possibilities, integrating multiple scientific data sources with advanced streaming frameworks. Additionally, our framework effectively utilizes computing resources and bridges the gaps identified by HPC experts. Our proposed approach tackles scalability and persistence challenges by leveraging native persistency, which has posed difficulties in traditional approaches. Furthermore, we seek to enhance fault-tolerance and load-balance of computations by leveraging real-time streaming in diverse scientific computing environments, thereby propelling advanced scientific computing research into the next generation.
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.
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.
This report details the impact exascale will bring to turbulent-flow simulations in applied science and technology. The need for accurate simulation of turbulent flows is evident across the DOE applied-science and engineering portfolios, including combustion, plasma physics, nuclear-reactor physics, wind energy, and atmospheric science. The workshop brought together experts in turbulent-flow simulation, computational mathematics, and high-performance computing. Building upon previous ASCR workshops on exascale computing, participants defined a research agenda and path forward that will enable scientists and engineers to continually leverage, engage, and direct advances in computational systems on the path to exascale computing.
Mixed-precision algorithms have been proposed as a way for scientific computing to benefit from some of the gains seen for AI on recent high performance computing (HPC) platforms. A few applications dominated by dense matrix operations have seen substantial speedups by utilizing low precision formats such as FP16. However, a majority of scientific simulation applications are memory bandwidth limited. Beyond preliminary studies, the practical gain from using mixed-precision algorithms on a given high-performance computing (HPC) system is largely unclear. The High Performance GMRES Mixed Precision (HPG-MxP) benchmark has been proposed to measure the useful performance of a HPC system on sparse matrix-based mixed-precision applications. In this work, we present an implementation of the HPG-MxP benchmark for an exascale system and describe our algorithm enhancements. We show for the first time a speedup of 1.6x using a combination of double- and single-precision keeping the same residual level on modern GPU-based supercomputers.
The design of high-performance computing architectures demands performance analysis of large-scale parallel applications to derive various parameters concerning hardware design and software development. The process of performance analysis and benchmarking an application can be done in several ways with varying degrees of fidelity. One of the most cost-effective ways is to do a coarse-grained study of large-scale parallel applications through the use of program skeletons. The concept of a “program skeleton” that we discuss in this article is an abstracted program that is derived from a larger program where source code that is determined to be irrelevant is removed for the purposes of the skeleton. In this work, we develop a semiautomatic approach for extracting program skeletons based on compiler program analysis. Finally, we demonstrate correctness of our skeleton extraction process by comparing details from communication traces, as well as show the performance speedup of using skeletons by running simulations in the SST/macro simulator.
The computation of the spin-rotation tensor within the framework of auxiliary density functional theory (ADFT) in combination with the gauge including atomic orbital (GIAO) scheme, to treat the gauge origin problem, is presented. For the spin-rotation tensor, the calculation of the magnetic shielding tensor represents the most demanding computational task. Employing the ADFT-GIAO methodology, the central processing unit time for the magnetic shielding tensor calculation can be dramatically reduced. In this work, the quality of spin-rotation constants obtained with the ADFT-GIAO methodology is compared with available experimental data as well as with other theoretical results at the Hartree-Fock and coupled-cluster level of theory. It is found that the agreement between the ADFT-GIAO results and the experiment is good and very similar to the ones obtained by the coupled-cluster single-doubles-perturbative triples-GIAO methodology. With the improved computational performance achieved, the computation of the spin-rotation tensors of large systems or along Born-Oppenheimer molecular dynamics trajectories becomes feasible in reasonable times. Three models of carbon fullerenes containing hundreds of atoms and thousands of basis functions are used for benchmarking the performance. Furthermore, a theoretical study of temperature effects on the structure and spin-rotation tensor of the H{sup 12}C–{sup 12}CH–DF complex is presented. Here, the temperature dependency of the spin-rotation tensor of the fluorine nucleus can be used to identify experimentally the so far unknown bent isomer of this complex. To the best of our knowledge this is the first time that temperature effects on the spin-rotation tensor are investigated.
Combustion encompasses multiscale, multiphase reacting flow physics spanning a wide range of scales from the molecular scales, where chemical reactions occur, to the device scales, where the turbulent flow is affected by the geometry of the combustor. This scale disparity and the limited measurement capabilities from experiments make modeling combustion a significant challenge. Recent advancements in high-performance computing (HPC), particularly with the Department of Energy's Exascale Computing Project (ECP), have enabled high-fidelity simulations of practical applications to be performed. The major physics submodels, including chemical reactions, turbulence, sprays, soot, and thermal radiation, exhibit distinctive computational characteristics that need to be examined separately to ensure efficient utilization of computational resources. This paper presents the multiphysics solver for the Pele code suite, called PeleMP, which consists of models for spray, soot, and thermal radiation. Here, the mathematical and algorithmic aspects of the model implementations are described in detail as well as the verification process. The computational performance of these models is benchmarked on multiple supercomputers, including Frontier, an exascale machine. Results are presented from production simulations of a turbulent sooting ethylene flame and a bluff-body swirl stabilized spray flame with sustainable aviation fuels to demonstrate the capability of the Pele codes for modeling practical combustion problems with multiphysics. This work is an important step toward the exascale computing era for high-fidelity combustion simulations providing physical insights and data for predictive modeling of real-world devices.
We design resource management heuristics that assign serial tasks to the nodes of a heterogeneous high performance computing (HPC) system. The value of completing these tasks is modeled using monotonically decreasing utility functions that represent the time-varying importance of the task. The value of completing a task is equal to its utility function at the time of its completion. The overall performance of this system is measured using the total utility earned by all tasks during some interval of time. To maximize the performance of such a system where the preemption of tasks is possible, we have designed, analyzed, and compared a set of resource allocation heuristic techniques. We combine two utility-aware heuristics with three different preemption techniques to create six preemption-capable heuristics. We also consider the two utility-aware heuristics without preemption. We use simulation studies to evaluate this set of eight heuristics and compare them with an FCFS heuristic, which is often used in real systems, and random assignments. In general, our set of eight heuristics is able to significantly outperform the comparison heuristics, and the preemption-capable heuristics are able to significantly increase the utility earned compared to the heuristics that do not use preemption. We analyze the performance tradeoffs among the different preemption-capable heuristics under a variety of oversubscribed workload environments.
Synchrophasor technology is widely deployed in the energy management system to monitor the grid health at micro level and perform necessary corrective actions in real time; however, integrated phasor devices and data aggregators are exposed to several cybersecurity threats. This paper proposes a federated ML(FML)-based ADS to detect several data integrity attacks in the synchrophasor network. The proposed approach integrates the horizontal FML technique and consists of substation-based local models and a control center-based global model. The proposed methodology includes training local models using heterogeneous data sets that include network and grid information and updating the global model through multiple iterations by sharing model gradients. Finally, the trained global model is applied to identify cyberattacks, normal operation, and physical events. To validate the proof of concept, we used synthetic data sets generated by Mississippi State University and Oak Ridge National Laboratory for training and testing the classification models using the National Renewable Energy Laboratory's high performance computing resources. Our experimental results, computed through several performance measures, reveal that the proposed approach shows consistent performance during the binary, three-class, and multiclass classifications while ensuring privacy of synchrophasor data.
This paper describes the application of a large-scale active learning method to characterize the parameter space of a computational agent-based model developed to investigate the impact of CommunityRx, a clinical information-based health intervention that provides patients with personalized information about local community resources to meet basic and self-care needs. Additionally, the diffusion of information about community resources and their use is modeled via networked interactions and their subsequent effect on agents' use of community resources across an urban population. A random forest model is iteratively fitted to model evaluations to characterize the model parameter space with respect to observed empirical data. We demonstrate the feasibility of using high-performance computing and active learning model exploration techniques to characterize large parameter spaces; by partitioning the parameter space into potentially viable and non-viable regions, we rule out regions of space where simulation output is implausible to observed empirical data. We argue that such methods are necessary to enable model exploration in complex computational models that incorporate increasingly available micro-level behavior data. We provide public access to the model and high-performance computing experimentation code.
Following the mass adoption of external accelerators for high performance computing, the overall performance of many applications has become increasingly dependent on relatively small accelerated kernels. As static analysis is fundamentally limited by dynamic values and external definitions, standard ahead-of-time compilation is not always sufficient to achieve the best performance. Furthermore, many users looking to port an existing application to run on an external accelerator will not want to fundamentally restructure their programs. These and other problems can be addressed through both link-time optimization (LTO) and just-in-time (JIT) compilation, but until now had sparse and inconsistent support from the compiler. In this work, we present a new compilation method that enables device-side LTO as well as a transparent JIT compilation tool-chain for OpenMP target offloading. Our contributions include an entirely new device linking and embedding scheme to enable LTO as well as a novel JIT engine to efficiently optimize OpenMP offloading regions at run-time. We also introduce a persistent caching system to improve end-to-end runtime using the JIT engine and minimize kernel launching overheads. We measure the performance of our LTO and JIT implementation via several real-world scientific applications. With our optimizations we observe significant improvements through LTO on large applications as well as significant end-to-end execution time improvement using JIT.
Here, this paper presents a comprehensive software stack architecture for integrating quantum computing (QC) capabilities with High-Performance Computing (HPC) environments. While quantum computers show promise as specialized accelerators for scientific computing, their effective integration with classical HPC systems presents significant technical challenges. We propose a hardware-agnostic software framework that supports both current noisy intermediate-scale quantum devices and future fault-tolerant quantum computers, while maintaining compatibility with existing HPC workflows. The architecture includes a quantum gateway interface, standardized APIs for resource management, and robust scheduling mechanisms to handle both simultaneous and interleaved quantum–classical workloads. Key innovations include: (1) a unified resource management system that efficiently coordinates quantum and classical resources, (2) a flexible quantum programming interface that abstracts hardware-specific details, (3) A Quantum Platform Manager API that simplifies the integration of various quantum hardware systems, and (4) a comprehensive tool chain for quantum circuit optimization and execution. We demonstrate our architecture through implementation of quantum–classical algorithms, including the variational quantum linear solver, showcasing the framework’s ability to handle complex hybrid workflows while maximizing resource utilization. This work provides a foundational blueprint for integrating QC capabilities into existing HPC infrastructures, addressing critical challenges in resource management, job scheduling, and efficient data movement between classical and quantum resources.
In recent years, advanced and complex analysis workflows have gained increasing importance in the ATLAS experiment at CERN, one of the large scientific experiments at LHC. Support for such workflows has allowed users to exploit remote computing resources and service providers distributed worldwide, overcoming limitations on local resources and services. The spectrum of computing options keeps increasing across the Worldwide LHC Computing Grid (WLCG), volunteer computing, high-performance computing, commercial clouds, and emerging service levels like Platform-as-a-Service (PaaS), Container-as-a-Service (CaaS) and Function-as-a-Service (FaaS), each one providing new advantages and constraints. Users can significantly benefit from these providers, but at the same time, it is cumbersome to deal with multiple providers, even in a single analysis workflow with fine-grained requirements coming from their applications’ nature and characteristics. In this paper, we will first highlight issues in geographically-distributed heterogeneous computing, such as the insulation of users from the complexities of dealing with remote providers, smart workload routing, complex resource provisioning, seamless execution of advanced workflows, workflow description, pseudointeractive analysis, and integration of PaaS, CaaS, and FaaS providers. We will also outline solutions developed in ATLAS with the Production and Distributed Analysis (PanDA) system and future challenges for LHC Run4.