Search NASA⌕ Search

SEARCH · Search NASA

Results for “high performance computing systems”

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 73 records · Page 4

Domaine-Specific Runtime to Orchestrate Computation on Heterogeneous Platforms

Task-based runtime systems in the past have sought to exploit inherent asynchronicity in the application execution to reduce overall runtime. In the last decade focus shifted to supporting the heterogeneity that is increasingly prevalent in high-performance computing systems. However, existing task-based runtime systems, being general, come with a challenging set of issues such as the complexity of abstractions and overheads. And they still leave much of the burden of exposing the parallelism on the application developers who also have to fit their applications to the runtime systems’ interfaces. In this paper we take a different approach for applications to target heterogeneous systems through domain-specific run-times. Our design aspires to leverage the domain-specific knowledge of a focused class of scientific simulations to pragmatically orchestrate computations in the simulations.

97 MATHEMATICS AND COMPUTING↗

A scalable framework for efficient coupling of thermal and microstructural simulations in additive manufacturing

Predicting microstructure evolution in metal additive manufacturing (AM) is important for process optimization, but spatiotemporal scale disparities between thermal transport and microstructure evolution create significant challenges for efficient data transfer between simulation codes. To address this, we present Stork, a scalable framework for coupling thermal and microstructural simulations. Stork uses a sparse data representation to identify and store active solidification sub-volumes, enabling highly parallel quad-linear interpolation from coarse thermal grids to fine microstructure grids without large intermediate storage. We demonstrate the framework by coupling the semi-analytic heat transfer code 3DThesis with the time-parallel cellular automata code Toucan. This approach achieves over two orders of magnitude reduction in data generation time and file size compared to prior workflows. Numerical studies show that quad-linear interpolation preserves grain morphology and crystallographic texture in laser powder bed fusion (LPBF) simulations for coarsening ratios up to 16. Overall, Stork provides a scalable pathway for high-throughput, component-scale AM simulations on modern high-performance computing systems.

36 MATERIALS SCIENCE↗

Large-scale GW calculations on pre-exascale HPC systems

The ab initio GW approach is a rigorous Green’s-function-based framework that can be employed to compute electronic excitation properties of a wide variety of materials such as extended systems, molecules, as well as confined and nanostructured materials with a very satisfactory accuracy. However, GW calculations on complex systems are often hindered by the high computational cost associated with the method. Here, in this study, we demonstrate how to significantly speedup GW calculations with a novel algorithm for the computationally intense kernel based on a non-blocking chunked cyclic communication scheme that minimizes latency in MPI messages, allows for overlapping of communication and computation, and improves cache usage in matrix multiplication operations. The optimized version of the code, implemented in the BerkeleyGW software package, is capable of scaling well to the full Cori computer at NERSC (Cray XC40) and achieves over 11 Peta FLOP/s of sustained performance. We showcase our work by performing large-scale GW calculations of defect structures on silicon, which require simulation cells containing over 1700 atoms, and which can now be efficiently executed in just a few minutes on large pre-exascale high-performance computing systems.

97 MATHEMATICS AND COMPUTING↗

Neuromorphic Graph Algorithms: Cycle Detection, Odd Cycle Detection, and Max Flow

Neuromorphic computing is poised to become a promising computing paradigm in the post Moore’s law era due to its extremely low power usage and inherent parallelism. Spiking neural networks are the traditional use case for neuromorphic systems, and have proven to be highly effective at machine learning tasks such as control problems. More recently, neuromorphic systems have been applied outside of the arena of machine learning, primarily in the field of graph algorithms. Neuromorphic systems have been shown to perform graph algorithms faster and with lower power consumption than their traditional (GPU/CPU) counterparts, and are hence an attractive option for a co-processing unit in future high performance computing systems, where graph algorithms play a critical role. In this paper, we present a neuromorphic implementation of cycle detection, odd cycle detection, and the Ford-Fulkerson max-flow algorithm. We further evaluate the performance of these implementations using the NEST neuromorphic simulator by using spike counts and simulation time as proxies for energy consumption and run time. In addition to gains inherent in neuromorphic systems, we show that within the neuromorphic implementations early stopping criteria can be implemented to further improve performance.

Kay, Bill↗

Adiak

Adiak is a computer science data management interface tool for collecting metadata about an application's runs and disseminating it to tools that subscribe to this data. Examples of this metadata could be what user is running an application, what systems it runs on, or application provided name/value pairs. Adiak's capabilities include automated reading of system-level metadata, and a generic type system for applications to register their own name/value metadata. Adiak is currently designed to provide a standard interface that can be integrated with MPI-based applications that run on high performance computing systems. Example tools that use Adiak metadata include performance analysis tools and workflow tools. By allowing the application to specify metadata to Adiak, which then disseminates it to tools, application developers do not have to write to tool-specific interfaces.

LeGendre, MatthewP↗

Comparative evaluation of deep learning workloads for leadership-class systems

Deep learning (DL) workloads and their performance at scale are becoming important factors to consider as we design, develop and deploy next-generation high-performance computing systems. Since DL applications rely heavily on DL frameworks and underlying compute (CPU/GPU) stacks, it is essential to gain a holistic understanding from compute kernels, models, and frameworks of popular DL stacks, and to assess their impact on science-driven, mission-critical applications. At Oak Ridge Leadership Computing Facility (OLCF), we employ a set of micro and macro DL benchmarks established through the Collaboration of Oak Ridge, Argonne, and Livermore (CORAL) to evaluate the AI readiness of our next-generation supercomputers. In this paper, we present our early observations and performance benchmark comparisons between the Nvidia V100 based Summit system with its CUDA stack and an AMD MI100 based testbed system with its ROCm stack. We take a layered perspective on DL benchmarking and point to opportunities for future optimizations in the technologies that we consider.

Yin, Junqi↗

A Hardware and Software Co-design Framework for Energy Efficient Neuromorphic Systems

Neuromorphic systems can be realized by a variety of algorithms and architectures. A common understanding is that spiking neuromorphic designs, which encode information into spatio-temporal spiking events, are both a biologically-accurate and efficient way of processing information. However, representing the information through timing relationships induces sophisticated circuit designs in traditional CMOS-based implementations. In recent years, high-capacity resistive memory (RRAM, aka, memristor) has demonstrated great potential in mimicking synaptic behaviors. Several RRAM-based spiking neuromorphic designs exist, most of which focus on rate coding schemes. These designs simplify circuit implementations of neuron models and explore challenges such as unsatisfactory speed, resolution, and performance. As an alternative, we will explore temporal coding spiking neuromorphic systems that encode information as the relative timing of neuron activations (spikes), which have been proven to be more adaptive and energy-efficient. Developing a neuromorphic system for spiking neural network (SNN) inference and online training, however, faces some major technical challenges: (1) It lacks circuit implementation support for temporal-coding SNN to achieve satisfying power efficiency and accuracy; (2) Although existing research works have investigated memristive synapse and neuron designs for spike-timing-dependent plasticity, the non-ideal conditions in implementation, such as device variations and signal degradation, degrade online learning accuracy of large scale systems; and (3) Non-optimized, inter-layer data traffic in SNNs, leads to unnecessary data communication costs. In this project, we plan to address these challenges by a hardware and software co-design framework that incorporates solutions at the circuit, architecture, and algorithm levels. At the circuit-level, we will elaborate on the in-situ SNN processing element designs for supporting both inference and online training modes. Variation-aware schemes will be studied to improve reliability. At the architecture level, we propose a pipelined, asynchronous architecture to retain the timing resolution of spikes. At the algorithm level, we will investigate an innovative SNN training algorithm for enabling activation sparsification and reducing unnecessary data communication costs. This neuromorphic system will provide an effective solution to real-life energy-constrained applications and significantly contribute to the exploration of next-generation high-performance computing systems under the DOE context.

97 MATHEMATICS AND COMPUTING↗

Improving Progressive Retrieval for HPC Scientific Data using Deep Neural Network

As the disparity between compute and I/O on high-performance computing systems has continued to widen, it has become increasingly difficult to perform post-hoc data analytics on full-resolution scientific simulation data due to the high I/O cost. Error-bounded data decomposition and progressive data retrieval framework has recently been developed to address such a challenge by performing data decomposition before storage and reading only part of the decomposed data when necessary. However, the performance of the progressive retrieval framework has been suffering from the over-pessimistic error control theory, such that the achieved maximum error of recomposed data is significantly lower than the required error. Therefore, more data than required is fetched for recomposition, incurring additional I/O overhead. In order to tackle this issue, we propose a DNN-based progressive retrieval framework that can better identify the minimum amount of data to be retrieved. Our contributions are as follows: 1) We provide an in-depth investigation of the recently developed progressive retrieval framework; 2) We propose two designs of prediction models (named D-MGARD and E-MGARD) to estimate the amount of retrieved data size based on error bounds. 3) We evaluate our proposed solutions using scientific datasets generated by real-world simulations from two domains. Evaluation results demonstrate the effectiveness of our solution in accurately predicting the amount of retrieval data size, as well as the advantages of our solution over the traditional approach to reducing the I/O overhead. Based on our evaluation, our solution is shown to read significantly less data (5% - 40% with D-MGARD, 20% - 80% with E-MGARD).

Wang, Jinzhen↗

Designing a parallel Feel-the-Way clustering algorithm on HPC systems

This paper introduces a new parallel clustering algorithm, named Feel-the-Way clustering algorithm, that provides better or equivalent convergence rate than the traditional clustering methods by optimizing the synchronization and communication costs. Our algorithm design centers on how to optimize three factors simultaneously: reduced synchronizations, improved convergence rate, and retained same or comparable optimization cost. To compare the optimization cost, we use the Sum of Square Error (SSE) cost as the metric, which is the sum of the square distance between each data point and its assigned clusters. Compared with the traditional MPI k-means algorithm, the new Feel-the-Way algorithm requires less communications among participating processes. As for the convergence rate, the new algorithm requires fewer number of iterations to converge. As for the optimization cost, it obtains the SSE costs that are close to the k-means algorithm. In the paper, we first design the full-step Feel-the-Way k-means clustering algorithm that can significantly reduce the number of iterations that are required by the original k-means clustering method. Next, we improve the performance of the full-step algorithm by adopting an optimized sampling-based approach, named reassignment-history-aware sampling. Our experimental results show that the optimized sampling-based Feel-the-Way method is significantly faster than the widely used k-means clustering method, and can provide comparable optimization costs. More extensive experiments with several synthetic datasets and real-world datasets (e.g., MNIST, CIFAR-10, ENRON, and PLACES-2) show that the new parallel algorithm can outperform the open source MPI k-means library by up to 110% on a high-performance computing system using 4,096 CPU cores. In addition, the new algorithm can take up to 51% fewer iterations to converge than the k-means clustering algorithm.

97 MATHEMATICS AND COMPUTING↗

Development and Experimental Optimization of High-Temperature Modeling Tools and Methods for Concentrated Solar Power Particle - Systems

A novel, open-source radiative modeling toolset was developed to extend the functionality of particle-based modeling software (e.g. discrete element method (DEM)) to environmental conditions relevant to concentrated solar power applications. This toolset was optimized for deployment on desktop workstations instead of high-performance computing systems, to render such tools more accessible to the research community. Both particle-based modeling and radiative exchange modeling are computationally expensive and often require specialized programming expertise, making these methods cumbersome to use. Recent developments in DEM software by DCS Computing have greatly reduced these challenges, providing a graphical-user-interface based platform and modeling optimization for desktop workstations, HPCs, and cloud computing. The University of Dayton leveraged the experience of DCS Computing in developing a user-friendly, open-source radiative heat transfer expansion for DEM modeling. The University of Dayton DEM+ radiative modeling toolset was developed using a combination of fundamental experimental measurements, modeling, and simplified flow experiments over a range of temperatures and flow conditions. The toolset provides researchers with access to multiple radiative models including an accelerated Monte-Carlo Ray Tracing (application agnostic, highly computationally expensive), an expanded database of distance-based approximations (application limited, computationally light), and a weighted blending of the two methods capable of achieving over 90% reduction in computation time with equivalent accuracy compared to Monte-Carlo Ray Tracing. Through a graphical user interface, users can customize the radiative models to match their desired accuracy and available computational resources, improving access to particle based modeling for the research community. Ceramic sintered bauxite proppants were used in modeling and experimentally as a baseline. Both the radiative heat transfer and flow properties for particulate systems were investigated at elevated temperatures up to 800 °C. The major accomplishments for this work include a verified, open-source radiative modeling toolset to be distributed amongst the research community and the fabrication of three small-scale test facilities to investigate particle behavior and tune DEM flow properties for operation up to 800 °C. The findings have been shared with the research community via conference modeling workshops, deployment of the tools in DCS Computing Aspherix®, and open-source access to the developed radiative modeling tool. The development of next-generation CSP facilities and thermal energy storage systems based on ceramic particles requires providing access to computationally efficient and accurate modeling tools. Particles will experience a wide range of environments (20-800 °C) and handling conditions (dilute curtains or dense packing), requiring specially designed and optimized equipment. Optimizing solid particle physics models and establishing best-practices for particle modeling in CSP environments will assist researchers with designing optimized equipment, accelerating the deployment of more economically-competitive CSP facilities.

14 SOLAR ENERGY↗

A Performance Study of Quantum ESPRESSO’s PWscf Code on Multi-core and GPU Systems

We describe the porting of PWscf (Plane-Wave Self Consistent Field), a key component of the Quantum ESPRESSO open-source suite of codes for materials modeling, to GPU systems using CUDA Fortran. Kernel loop directives (CUF kernels) have been extensively used in order to have a single source code for both CPU and GPU implementations. The results of the GPU version have been carefully validated and the performance of the code on several GPU systems (both x86 and POWER8 based) has been compared with traditional Intel multi-core (CPU only) systems. This current GPU version can reduce the time-to-solution by an average factor of 2–3 running two different input cases widely used as benchmarks on small and large high performance computing systems.

CUDA Fortran↗

High-bandwidth density silicon photonic resonators for energy-efficient optical interconnects

The growth of artificial intelligence applications demands ever larger and more complex deep learning models, dominating today's—and tomorrow's—data center and high-performance computing systems. While traditional electronics are failing to keep pace with application demands, silicon photonic (SiPh) interconnects have emerged as a necessary technology to support these systems. SiPh-driven wavelength-division multiplexing (WDM) offers a particularly promising path toward supporting incredibly high-aggregate link bandwidth in a compact and efficient form factor. One of the basic building blocks of these integrated WDM interconnects is the SiPh resonator. Their inherent wavelength selectivity and compact footprint allow for efficient data transmission multiplexed across dozens of carrier wavelengths. Used as add-drop (AD) filters, SiPh resonators are critical to constructing integrated tunable wavelength-selective optical circuit switches as well as for demultiplexing the different carrier wavelengths toward independent wavelength-insensitive photodiodes in a dense wavelength-division multiplexing receiver. Resonators in the all-pass (AP) configuration are widespread as well, allowing for wavelength-selective modulation to drive aggregate link bandwidths far beyond the individual channel data rate. Unlike SiPh Mach–Zehnder modulators (MZM), resonant modulators can be driven using low, complementary metal-oxide-semiconductor drive voltages, allowing for tight co-integration between photonic integrated circuits, fabricated with larger process node technologies, and electronic integrated circuits, designed to exploit the advantages of the latest node. To push toward practical peta-scale interconnects, a comprehensive review of SiPh resonators is required, addressing bottlenecks and design constraints at both the architecture and device levels. We first describe the predominant integrated link architectures and identify their limits. We then discuss the device-level design considerations that can be made for both AD and AP configuration resonators to overcome the system level limits with novel resonator device designs. Analytical models and numerical simulation of resonators are validated by experimental measurement of devices fabricated in a commercial 300-mm foundry, showing a clear path toward volume manufacturing. The demonstrated resonant modulators and filters support the feasibility of increasing the aggregate bandwidth of resonator-driven SiPh interconnects into the peta-scale regime.

42 ENGINEERING↗

Enforcing End-to-end I/O Policies for Scientific Workflows using Software-Defined Storage Resource Enclaves

Data-intensive knowledge discovery requires scientific applications to run concurrently with analytics and visualization codes executing in situ for timely output inspection and knowledge extraction. Consequently, I/O pipelines of scientific workflows can be long and complex because they comprise many stages of analytics across different layers of the I/O stack of high-performance computing systems. Performance limitations at any I/O layer or stage can cause an I/O bottleneck resulting in greater than expected end-to-end I/O latency. In this paper, we present the design and implementation of a novel data management infrastructure called Software-Defined Storage Resource Enclaves (SIREN) at system level to enforce end-to-end policies that dictate an I/O pipeline's performance. SIREN provides an I/O performance interface for users to specify the desired storage resources in the context of in-situ analytics. If suboptimal performance of analytics is caused by an I/O bottleneck when data are transferred between simulations and analytics, schedulers in different layers of the I/O stack automatically provide the guaranteed lower bounds on I/O throughput. Lastly, our experimental results demonstrate that SIREN provides performance isolation among scientific workflows sharing multiple storage servers across two I/O layers (burst buffer and parallel file systems) while maintaining high system scalability and resource utilization.

97 MATHEMATICS AND COMPUTING↗

High Fidelity CFD Simulations Supporting the KP-FHR

Kairos Power, LLC, is developing its version of the Fluoride-cooled High-temperature Reactor, the KP-FHR. The design uses a pebble bed core with fluoride salt as a coolant. The pebbles used in the KP-FHR have a diameter of 4 cm, with a shell fuel region where TRISO particles are embedded. A Pebble bed core design is adopted by several Gen IV reactors, They boast many benefits, such as fuel integrity, highly efficient heat transfer, and passive safety. However, it is challenging to accurately predict temperature and flow inside a pebble bed. Traditional approaches use the porous media model, which regards the pebble bed as a continuous medium, but with different temperature fields representing different levels, such as the fluid temperature, pebble surface temperature, and pebble center temperature. Empirical heat transfer correlations are adopted to calculate the heat transfer coefficient between different phases. However, empirical correlations are usually validated with experimental data, which usually lacks detail inside the pebble bed. The available experimental data is also generally at a high Reynolds number, which falls outside of the conditions of KP-FHR. Explicit computational fluid dynamics (CFD) simulations of randomly packed pebble beds have only become feasible recently. This is thanks to the rapid development of computational power and scalable algorithms. In this work, we used the Spectral Element Method (SEM) CFD code NekRS to simulate the randomly packed pebble bed in a cylindrical container. NekRS, which is the GPU variant of Nek5000, but refactored to utilize the computational power of GPUs using the OCCA library to run on hybrid architecture high performance computing systems. It was initially developed with the libParamunal library, but truncated and tuned for large-scale turbulence simulation. As a result, the SEM reaches higher precision with the same degrees of freedom by using a high-order Lagrange polynomial basis distributed on Gauss-Lobatto-Legendre quadrature inside each element, compared to lower-order methods, such the Finite Volume Method and Finite Element Method. The report is divided into five parts. We start with a general discussion of the pebble bed reactor, along with a specific investigation into the KP-FHR. The second part presents the numerical methodology. In the third part, we study a modular pebble bed with 1741 pebbles in a container of 7 pebble-diameter radius. Beyond LES simulations done by NekRS, we also leveraged the thermal radiation model in OpenFOAM to study heat transfer under no-forced-flow scenarios. Then, in the fourth part we simulated a pebble bed similar to the size of the Hermes Test Reactor. The total number of pebbles is in these simulations is 34,374. The container radius is 14 pebble-diameters. Finally, the report concludes in part five, with a discussion of future work.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

OpenMP Target Task: Tasking and Target Offloading on Heterogeneous Systems

This work evaluated the use of OpenMP tasking with target GPU offloading as a potential solution for programming productivity and performance on heterogeneous systems. Also, it is proposed a new OpenMP specification to make the implementation of heterogeneous codes simpler by using OpenMP target task, which integrates both OpenMP tasking and target GPU offloading in a single OpenMP pragma. As a test case, the authors used one of the most popular and widely used Basic Linear Algebra Subprogram Level-3 routines: triangular solver (TRSM). To benefit from the heterogeneity of the current high-performance computing systems, the authors propose a different parallelization of the algorithm by using a nonuniform decomposition of the problem. This work used target GPU offloading inside OpenMP tasks to address the heterogeneity found in the hardware. This new approach can outperform the state-of-the-art algorithms, which use a uniform decomposition of the data, on both the CPU-only and hybrid CPU-GPU systems, reaching speedups of up to one order of magnitude. The performance that this approach achieves is faster than the IBM ESSL math library on CPU and competitive relative to a highly optimized heterogeneous CUDA version. One node of Oak Ridge National Laboratory’s supercomputer, Summit, was used for performance analysis.

Valero Lara, Pedro↗

CP2K: An Electronic Structure and Molecular Dynamics Software Package - Quickstep: Efficient and Accurate Electronic Structure Calculations

CP2K is an open source electronic structure and molecular dynamics software package to perform atomistic simulations of solid-state, liquid, molecular and biological systems. It is especially aimed at massively-parallel and linear-scaling electronic structure methods and state-of-the-art ab-initio molecular dynamics simulations. Excellent performance for electronic structure calculations is achieved using novel algorithms implemented for modern high-performance computing systems. This review revisits the main capabilities of CP2K to perform efficient and accurate electronic structure simulations. The emphasis is put on density functional theory and multiple post-Hartree-Fock methods using the Gaussian and plane wave approach and its augmented all-electron extension. TDK has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No. 716142). VRR has been supported by the Swiss National Science Foundation in the form of Ambizione grant No. PZ00P2 174227 and RZK by the Natural Sciences and Engineering Research Council of Canada (NSERC) through Discovery Grants (RGPIN-2016-0505). GKS and CJM are supported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences. UK based work was funded under the embedded CSE programme of the ARCHER UK National Supercomputing Service (http://www.archer.ac.uk), grants eCSE03-011, eCSE06-6, eCSE08-9, eCSE13-17 and the EPSRC (EP/P022235/1) grant “Surface and Interface Toolkit for the Materials Chemistry Community". Computational resources were provided by the Swiss National Supercomputing Centre (CSCS) and Compute Canada. The generous allocation of computing time on the FPGA-based supercomputer “Noctua" at PC2 is kindly acknowledged.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Scaling Structured Multigrid to 500K+ Cores Through Coarse-Grid Redistribution

The effcient solution of sparse, linear systems resulting from the discretization of partial differential equations is crucial to the performance of many physics-based simulations. The algorithmic optimality of multilevel approaches for common discretizations makes them a good candidate for an efficient parallel solver. Yet, modern architectures for high-performance computing systems continue to challenge the parallel scalability of multilevel solvers. While algebraic multigrid methods are robust for solving a variety of problems, the increasing importance of data locality and cost of data movement in modern architectures motivates the need to carefully exploit structure in the problem. Robust logically structured variational multigrid methods, such as Black Box Multigrid (BoxMG), maintain structure throughout the multigrid hierarchy. This avoids indirection and increased coarse- grid communication costs typical in parallel algebraic multigrid. Nevertheless, the parallel scalability of structured multigrid is challenged by coarse-grid problems where the overhead in communication dominates computation. In this paper, an algorithm is introduced for redistributing coarse-grid problems through incremental agglomeration. Guided by a predictive performance model, this algorithm provides robust redistribution decisions for structured multilevel solvers. A two-dimensional diffusion problem is used to demonstrate the significant gain in performance of this algorithm over the previous approach that used agglomeration to one processor. In addition, the parallel scalability of this approach

97 MATHEMATICS AND COMPUTING↗