Search NASA⌕ Search

SEARCH · Search NASA

Results for “accelerated computing”

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 955 records · Page 53

Implementing Molecular Dynamics on Hybrid High Performance Computers - Three-Body Potentials

The use of coprocessors or accelerators such as graphics processing units (GPUs) has become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power re- quirements. Hybrid high-performance computers, defined as machines with nodes containing more than one type of floating-point processor (e.g. CPU and GPU), are now becoming more prevalent due to these advantages. Although there has been extensive research into methods to efficiently use accelerators to improve the performance of molecular dynamics (MD) employing pairwise potential energy models, little is reported in the literature for models that include many-body effects. 3-body terms are required for many popular potentials such as MEAM, Tersoff, REBO, AIREBO, Stillinger-Weber, Bond-Order Potentials, and others. Because the per-atom simulation times are much higher for models incorporating 3-body terms, there is a clear need for efficient algo- rithms usable on hybrid high performance computers. Here, we report a shared-memory force-decomposition for 3-body potentials that avoids memory conflicts to allow for a deterministic code with substantial performance improvements on hybrid machines. We describe modifications necessary for use in distributed memory MD codes and show results for the simulation of water with Stillinger-Weber on the hybrid Titan supercomputer. We compare performance of the 3-body model to the SPC/E water model when using accelerators. Finally, we demonstrate that our approach can attain a speedup of 5.1 with acceleration on Titan for production simulations to study water droplet freezing on a surface.

Brown, W Michael↗

Divide and Conquer (DC) BLAST: fast and easy BLAST execution within HPC environments

Bioinformatics is currently faced with very large-scale data sets that lead to computational jobs, especially sequence similarity searches, that can take absurdly long times to run. For example, the National Center for Biotechnology Information (NCBI) Basic Local Alignment Search Tool (BLAST and BLAST+) suite, which is by far the most widely used tool for rapid similarity searching among nucleic acid or amino acid sequences, is highly central processing unit (CPU) intensive. While the BLAST suite of programs perform searches very rapidly, they have the potential to be accelerated. In recent years, distributed computing environments have become more widely accessible and used due to the increasing availability of high-performance computing (HPC) systems. Therefore, simple solutions for data parallelization are needed to expedite BLAST and other sequence analysis tools. However, existing software for parallel sequence similarity searches often requires extensive computational experience and skill on the part of the user. In order to accelerate BLAST and other sequence analysis tools, Divide and Conquer BLAST (DCBLAST) was developed to perform NCBI BLAST searches within a cluster, grid, or HPC environment by using a query sequence distribution approach. Scaling from one (1) to 256 CPU cores resulted in significant improvements in processing speed. Thus, DCBLAST dramatically accelerates the execution of BLAST searches using a simple, accessible, robust, and parallel approach. DCBLAST works across multiple nodes automatically and it overcomes the speed limitation of single-node BLAST programs. DCBLAST can be used on any HPC system, can take advantage of hundreds of nodes, and has no output limitations. Thus, this freely available tool simplifies distributed computation pipelines to facilitate the rapid discovery of sequence similarities between very large data sets.

59 BASIC BIOLOGICAL SCIENCES↗

Application of the Reduction of Scale Range in a Lorentz Boosted Frame to the Numerical Simulation of Particle Acceleration Devices

It has been shown [1] that it may be computationally advantageous to perform computer simulations in a boosted frame for a certain class of systems: particle beams interacting with electron clouds, free electron lasers, and laser-plasma accelerators. However, even if the computer model relies on a covariant set of equations, it was also pointed out that algorithmic difficulties related to discretization errors may have to be overcome in order to take full advantage of the potential speedup [2] . In this paper, we focus on the analysis of the complication of data input and output in a Lorentz boosted frame simulation, and describe the procedures that were implemented in the simulation code Warp[3]. We present our most recent progress in the modeling of laser wakefield acceleration in a boosted frame, and describe briefly the potential benefits of calculating in a boosted frame for the modeling of coherent synchrotron radiation.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Second order time evolution of the multigroup diffusion and P{sub 1} equations for radiation transport

Highlights: {yields} An existing multigroup transport algorithm is extended to be second-order in time. {yields} A new algorithm is presented that does not require a grey acceleration solution. {yields} The two algorithms are tested with 2D, multi-material problems. {yields} The two algorithms have comparable computational requirements. - Abstract: An existing solution method for solving the multigroup radiation equations, linear multifrequency-grey acceleration, is here extended to be second order in time. This method works for simple diffusion and for flux-limited diffusion, with or without material conduction. A new method is developed that does not require the solution of an averaged grey transport equation. It is effective solving both the diffusion and P{sub 1} forms of the transport equation. Two dimensional, multi-material test problems are used to compare the solution methods.

97 MATHEMATICS AND COMPUTING↗

A Generic Advanced Computing Framework for Executing Windows-based Dynamic Contingency Analysis Tool in Parallel on Cluster Machines

Dynamic contingency analysis tool (DCAT) has been developed to assess the impact and likelihood of extreme contingencies and potential cascading events across their systems and interconnections. By including more customized protection models and corrective actions into the Windows-based commercial tools, DCAT can help operators understand the cascading behavior and find mitigation approaches to reduce the risk of cascading outages in a more realistic manner. In order to further enhance the capability of DCAT, this paper presents a design of an advanced computing framework that enables DCAT to run on a cluster machine to improve its computational performance. This framework is generic and can be applied to other Windows-based simulation tools to fill the technical gap of applying advanced computing technology to vendors' Windows-based tools. The preliminary tests using medium to large power systems have shown the effectiveness of this framework and its potential for accelerating the adoption of advanced computing in utilities.

Advanced computing, dynamic contingency analysis, ↗

NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations

In the last 50 years there has been a tremendous progress in solving numerically the Navier-Stokes equations using finite differences, finite elements, spectral, and even meshless methods. Yet, in many real cases, we still cannot incorporate seamlessly (multi-fidelity) data into existing algorithms, and for industrial-complexity applications the mesh generation is time consuming and still an art. Moreover, solving ill-posed problems (e.g., lacking boundary conditions) or inverse problems is often prohibitively expensive and requires different formulations and new computer codes. Here, we employ physics-informed neural networks (PINNs), encoding the governing equations directly into the deep neural network via automatic differentiation, to overcome some of the aforementioned limitations for simulating incompressible laminar and turbulent flows. We develop the Navier-Stokes flow nets (NSFnets) by considering two different mathematical formulations of the Navier-Stokes equations: the velocity-pressure (VP) formulation and the vorticity-velocity (VV) formulation. Since this is a new approach, we first select some standard benchmark problems to assess the accuracy, convergence rate, computational cost and flexibility of NSFnets; analytical solutions and direct numerical simulation (DNS) databases provide proper initial and boundary conditions for the NSFnet simulations. The spatial and temporal coordinates are the inputs of the NSFnets, while the instantaneous velocity and pressure fields are the outputs for the VP-NSFnet, and the instantaneous velocity and vorticity fields are the outputs for the VV-NSFnet. This is unsupervised learning and, hence, no labeled data are required beyond boundary and initial conditions and the fluid properties. The residuals of the VP or VV governing equations, together with the initial and boundary conditions, are embedded into the loss function of the NSFnets. No data is provided for the pressure to the VP-NSFnet, which is a hidden state and is obtained via the incompressibility constraint without extra computational cost. Unlike the traditional numerical methods, NSFnets inherit the properties of neural networks (NNs), hence the total error is composed of the approximation, the optimization, and the generalization errors. Here, we empirically attempt to quantify these errors by varying the sampling (“residual”) points, the iterative solvers, and the size of the NN architecture. For the laminar flow solutions, we show that both the VP and the VV formulations are comparable in accuracy but their best performance corresponds to different NN architectures. The initial convergence rate is fast but the error eventually saturates to a plateau due to the dominance of the optimization error. For the turbulent channel flow, we show that NSFnets can sustain turbulence at , but due to expensive training we only consider part of the channel domain and enforce velocity boundary conditions on the subdomain boundaries provided by the DNS data base. We also perform a systematic study on the weights used in the loss function for balancing the data and physics components, and investigate a new way of computing the weights dynamically to accelerate training and enhance accuracy. In the last part, we demonstrate how NSFnets should be used in practice, namely for ill-posed problems with incomplete or noisy boundary conditions as well as for inverse problems. We obtain reasonably accurate solutions for such cases as well without the need to change the NSFnets and at the same computational cost as in the forward well-posed problems. As a result, we also present a simple example of transfer learning that will aid in accelerating the training of NSFnets for different parameter settings.

97 MATHEMATICS AND COMPUTING↗

Design and Implementation of the CEBAF Element Database

With inauguration of the CEBAF Element Database (CED) in Fall 2010, Jefferson Lab computer scientists have taken a first step toward the eventual goal of a model-driven accelerator. Once fully populated, the database will be the primary repository of information used for everything from generating lattice decks to booting front-end computers to building controls screens. A particular requirement influencing the CED design is that it must provide consistent access to not only present, but also future, and eventually past, configurations of the CEBAF accelerator. To accomplish this, an introspective database schema was designed that allows new elements, element types, and element properties to be defined on-the-fly without changing table structure. When used in conjunction with the Oracle Workspace Manager, it allows users to seamlessly query data from any time in the database history with the exact same tools as they use for querying the present configuration. Users can also check-out workspaces and use them as staging areas for upcoming machine configurations. All Access to the CED is through a well-documented API that is translated automatically from original C++ into native libraries for script languages such as perl, php, and TCL making access to the CED easy and ubiquitous. Notice: Authored by Jefferson Science Associates, LLC under U.S. DOE Contract No. DE-AC05-06OR23177. The U.S. Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce this manuscript for U.S. Government purposes.

43 PARTICLE ACCELERATORS↗

Design of Linear Accelerator (LINAC) tanks for proton therapy via Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) approaches

A homemade computer code for designing a Side- Coupled Linear Accelerator (SCL) is written. It integrates a simplified model of SCL tanks with the Particle Swarm Optimization (PSO) algorithm. The computer code main aim is to obtain useful guidelines for the design of Linear Accelerator (LINAC) resonant cavities. The design procedure, assisted via the aforesaid approach seems very promising, allowing future improvements towards the optimization of actual accelerating geometries. (authors)

43 PARTICLE ACCELERATORS↗

Production and Acceleration of Antinuclei in Supernova Shockwaves

We compute the energy spectra of antideuterons ( d-bar ) and antihelium ( (He)-bar ) in cosmic rays (CRs) in a scenario where hadronic interactions inside supernova remnants (SNRs) can produce a diffusively shock-accelerated “source component” of secondary antinuclei. The key parameters that specify the SNR environment and the interstellar CR transport are tightly constrained with the new measurements provided by the AMS experiment on the B/C ratio and on the p-bar /p ratio. The best-fit models obtained from the two ratios are found to be inconsistent with each other, as the p-bar /p data require enhanced secondary production. Thus, we derive conservative (i.e., B/C-driven) and speculative ( p-bar /p-driven) upper limits to the SNR flux contributions for the d-bar and (He)-bar spectra in CRs, along with their standard secondary component expected from CR collisions in the interstellar gas. We find that the source component of antinuclei can be appreciable at kinetic energies above a few ∼10 GeV n{sup −1}, but it is always sub-dominant below a few GeV n{sup −1}, that is the energy window where dark matter (DM) annihilation signatures are expected to exceed the level of secondary production. We also find that the total (standard + SNR) flux of secondary d-bar and (He)-bar is tightly constrained by the data. Thus, the presence of interaction processes in SNRs does not critically affect the total background for DM searches.

79 ASTRONOMY AND ASTROPHYSICS↗

Computation of absolute hydration and binding free energy with free energy perturbation distributed replica-exchange molecular dynamics.

Distributed Replica (REPDSTR) is a powerful parallelization technique enabling simulations of a group of replicas in a parallel/parallel fashion, where each replica is distributed to different nodes of a large cluster [Theor. Chem. Acc. 2003, 109, 140]. Here, we use the framework provided by REPDSTR to combine a staged free energy perturbation protocol with replica-exchange molecular dynamics (FEP/REMD). The structure of REPDSTR, which allows multiple parallel input/output (I/O), facilitates the treatment of replica-exchange to couple the N window simulations corresponding to different values of the thermodynamic coupling parameters. As a result, each of the N synchronous window simulations benefit from the sampling carried out by the N-1 others. As illustrative examples of the FEP/REMD strategy, calculations of the absolute hydration and binding free energy of small molecules were performed using the biomolecular simulation program CHARMM adapted for the IBM Blue Gene/P platform. The computations show that a FEP/REMD strategy significantly improves the sampling and accelerates the convergence of absolute free energy computations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Modeling laser-plasma acceleration in the wakefield frame

A simulation of laser-plasma acceleration in the boosted frame of the wake, moving at near lightspeed. Space has contracted and time has stretched, separating events in time. Relatively few time steps are needed to model them, requiring less computer time.

43 PARTICLE ACCELERATORS↗

Machine learning accelerated discrete element modeling of granular flows

Granular flows are widely encountered in many industrial processes and natural phenomena. Discrete Element Modeling (DEM) is a useful tool for understanding and troubleshooting devices, which handle granular materials. However, its applicability is significantly limited by the huge computational cost associated with detecting and computing collisions. In this research, the computation speed of DEM was accelerated by orders of magnitude using a convolutional neural network to replace the direct calculation of particle-particle and particle-boundary collisions. The MFiX software was used to generate the training and testing dataset. Additionally, a GPU accelerated TensorFlow model was used to train the neural network and test the results. The model fluctuations caused by different training steps were reduced with a multi-scale loss function. The accuracy was improved with more frames within one training step. The modeling of a rotating drum and a hopper demonstrated the accuracy and efficiency of this machine learning accelerated DEM in the simulation of granular flows.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

GPU-acceleration of tensor renormalization with PyTorch using CUDA

We show that numerical computations based on tensor renormalization group (TRG) methods can be significantly accelerated with PyTorch on graphics processing units (GPUs) by leveraging NVIDIA's Compute Unified Device Architecture (CUDA). Here we find improvement in the runtime and its scaling with bond dimension for two-dimensional systems. Our results establish that the utilization of GPU resources is essential for future precision computations with TRG.

97 MATHEMATICS AND COMPUTING↗

Challenges for the CMS computing model in the first year

CMS is in the process of commissioning a complex detector and a globally distributed computing infrastructure simultaneously. This represents a unique challenge. Even at the beginning there is not sufficient analysis or organized processing resources at CERN alone. In this presentation we discuss the unique computing challenges CMS expects to face during the first year of running and how they influence the baseline computing model decisions. During the early accelerator commissioning periods, CMS will attempt to collect as many events as possible when the beam is on in order to provide adequate early commissioning data. Some of these plans involve overdriving the Tier-0 infrastructure during data collection with recovery when the beam is off. In addition to the larger number of triggered events, there will be pressure in the first year to collect and analyze more complete data formats as the summarized formats mature. The large event formats impact the required storage, bandwidth, and processing capacity across all the computing centers. While the understanding of the detector and the event selections is being improved, there will likely be a larger number of reconstruction passes and skims performed by both central operations and individual users. We discuss how these additional stresses impact the allocation of resources and the changes from the baseline computing model.

99 GENERAL AND MISCELLANEOUS↗

Accelerated discovery of elpasolite scintillators

Elpasolite scintillators are a large family of halides which includes compounds reported to meet the NA22 program goals of <3% energy resolution at 662 keV1. This work investigated the potential to produce quality elpasolite compounds and alloys of useful sizes at reasonable cost, through systematic experimental and computational investigation of crystal structure and properties across the composition space. Discovery was accelerated by computational methods and models developed previously to efficiently identify cubic members of the elpasolite halides, and to evaluate stability of anion and cation exchange alloys.

36 MATERIALS SCIENCE↗

Efficient Heterogeneous Execution on Large Multicore and Accelerator Platforms: Case Study Using a Block Tridiagonal Solver

The algorithmic and implementation principles are explored in gainfully exploiting GPU accelerators in conjunction with multicore processors on high-end systems with large numbers of compute nodes, and evaluated in an implementation of a scalable block tridiagonal solver. The accelerator of each compute node is exploited in combination with multicore processors of that node in performing block-level linear algebra operations in the overall, distributed solver algorithm. Optimizations incorporated include: (1) an efficient memory mapping and synchronization interface to minimize data movement, (2) multi-process sharing of the accelerator within a node to obtain balanced load with multicore processors, and (3) an automatic memory management system to efficiently utilize accelerator memory when sub-matrices spill over the limits of device memory. Results are reported from our novel implementation that uses MAGMA and CUBLAS accelerator software systems simultaneously with ACML for multithreaded execution on processors. Overall, using 940 nVidia Tesla X2090 accelerators and 15,040 cores, the best heterogeneous execution delivers a 10.9-fold reduction in run time relative to an already efficient parallel multicore-only baseline implementation that is highly optimized with intra-node and inter-node concurrency and computation-communication overlap. Detailed quantitative results are presented to explain all critical runtime components contributing to hybrid performance.

Park, Alfred J↗

Scaling SQL to the Supercomputer for Interactive Analysis of Simulation Data

AI and simulation workloads consume and generate large amounts of data that need to be searched, transformed and merged with other data. With the goal of treating data as a first-class citizen inside a traditionally compute-centric HPC environment, we explore how the use of accelerators and high-speed interconnects can speed up tasks which otherwise constitute bottlenecks in computational discovery workflows. BlazingSQL is SQL engine that runs natively on NVIDIA GPUs and supports internode communication for fast analytics on terabyte-scale tabular data sets. We show how a fast interconnect improves query performance if leveraged through the Unified Communication X (UCX) middleware. We envision that future computing platforms will integrate accelerated database query capabilities for immediate and interactive analysis of large simulation data.

Glaser, Jens↗

The PetscSF Scalable Communication Layer

PetscSF, the communication component of the Portable, Extensible Toolkit for Scientific Computation (PETSc), is designed to provide PETSc's communication infrastructure suitable for exascale computers that utilize GPUs and other accelerators. PetscSF provides a simple application programming interface (API) for managing common communication patterns in scientific computations by using a star-forest graph representation. PetscSF supports several implementations based on MPI and NVSHMEM, whose selection is based on the characteristics of the application or the target architecture. An efficient and portable model for network and intra-node communication is essential for implementing large-scale applications. The Message Passing Interface, which has been the de facto standard for distributed memory systems, has developed into a large complex API that does not yet provide high performance on the emerging heterogeneous CPU-GPU-based exascale systems. Here, we discuss the design of PetscSF, how it can overcome some difficulties of working directly with MPI on GPUs, and we demonstrate its performance, scalability, and novel features.

97 MATHEMATICS AND COMPUTING↗