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 739 records · Page 41

Accelerating DCA++ (Dynamical Cluster Approximation) Scientific Application on the Summit supercomputer

Optimizing scientific applications on today’s accelerator-based high performance computing systems can be challenging, especially when multiple GPUs and CPUs with heterogeneous memories and persistent non-volatile memories are present. An example is Summit, an accelerator-based system at the Oak Ridge Leadership Computing Facility (OLCF) that is rated as the world’s fastest supercomputer to-date. New strategies are thus needed to expose the parallelism in legacy applications, while being amenable to efficient mapping to the underlying architecture.In this paper we discuss our experiences and strategies to port a scientific application, DCA++, to Summit. DCA++ is a highperformance research application that solves quantum manybody problems with a cutting edge quantum cluster algorithm, the dynamical cluster approximation.Our strategies aim to synergize the strengths of the different programming models in the code. These include: (a) streamlining the interactions between the CPU threads and the GPUs, (b) implementing computing kernels on the GPUs and decreasing CPU-GPU memory transfers, (c) allowing asynchronous GPU communications, and (d) increasing compute intensity by combining linear algebraic operations.Full-scale production runs using all 4600 Summit nodes attained a peak performance of 73.5 PFLOPS with a mixed precision implementation.We observed a perfect strong and weak scaling for the quantum Monte Carlo solver in DCA++, while encountering about 2× input/output (I/O) and MPI communication overhead on the time-to-solution for the full machine run. Our hardware agnostic optimizations are designed to alleviate the communication and I/O challenges observed, while improving the compute intensity and obtaining optimal performance on a complex, hybrid architecture like Summit.

Balduzzi, Giovanni↗

Lagrangian velocity and acceleration correlations of large inertial particles in a closed turbulent flow

We investigate the response of large inertial particle to turbulent fluctuations in an inhomogeneous and anisotropic flow. We conduct a Lagrangian study using particles both heavier and lighter than the surrounding fluid, and whose diameters are comparable to the flow integral scale. Both velocity and acceleration correlation functions are analyzed to compute the Lagrangian integral time and the acceleration time scale of such particles. The knowledge of how size and density affect these time scales is crucial in understanding particle dynamics and may permit stochastic process modelization using two-time models (for instance, Sawford’s). As particles are tracked over long times in the quasi-totality of a closed flow, the mean flow influences their behaviour and also biases the velocity time statistics, in particular the velocity correlation functions. By using a method that allows for the computation of turbulent velocity trajectories, we can obtain unbiased Lagrangian integral time. This is particularly useful in accessing the scale separation for such particles and to comparing it to the case of fluid particles in a similar configuration.

42 ENGINEERING↗

Using Hybrid Model OpenSHMEM + CUDA to Implement the SHOC Benchmark Suite

This work describes the process of porting the Scalable HeterOgeneous Computing (SHOC) benchmark suite from the hybrid MPI + CUDA implementation to OpenSHMEM + CUDA. SHOC includes a wide variety of benchmark kernels used to measure accelerator performance in both single node and cluster configurations. The hybrid model implementation attempts to place all major computation on accelerator devices, and uses MPI to synchronize and aggregate results. In some cases, MPI Groups are used to gradually reduce the number of accelerators used for computation as the problem size drops. Porting this behavior to OpenSHMEM required implementing several synchronizing collective operations, and using SHMEM teams to replace MPI Group functionality. Benchmark results on a Cray XK7 system with one GPU per compute node show that SHMEM performance is equal to MPI performance in these hybrid tasks. These results and porting experience show that using OpenSHMEM for accelerator devices benefits from adding functionality for synchronization and teams, and would further benefit from adding support for communication within accelerator kernels.

97 MATHEMATICS AND COMPUTING↗

Multi-scale observations of structure and chemical composition changes of portland cement systems during hydration

There is little agreement about the mechanisms and few direct observations of the transition of cement paste from a slurry to a solid. This paper uses four different imaging techniques at multiple length scales (from 15.6 nm to 1 μm) to follow the three-dimensional microstructural evolution of portland cement, and monoclinic tricalcium silicate (mC 3 S) paste over the first 16 h of hydration. Measurements of over 60,000 particles captured every 10 min were made at the micron scale. Nanoscale examinations of the structure and chemical composition are used to support this work. The results show that hydration products with an average Ca/Si > 3 form on and near the surface of the hydrating particles prior to the acceleration period that appears to control the reaction rate. These hydration products appear to change in chemical composition as more rapid dissolution of C 3 S and formation of hydration products occur during the acceleration period. The findings are used to provide insights into the origin of the induction and acceleration periods during hydration.

36 MATERIALS SCIENCE↗

Simulating Atmospheric Boundary Layer Turbulence with Nek5000/RS

We present large-eddy-simulation (LES) modeling approaches for the simulation of atmospheric boundary layer turbulence that are of direct relevance to wind energy production. In this report, we study a GABLS benchmark problem using high-order spectral element code Nek5000/RS, which is supported under the DOE’s Exascale Computing Project (ECP) Center for Efficient Exascale Discretizations (CEED) project, targeting application simulations on various acceleration-device based exascale computing platforms [1, 2]. We demonstrate our newly developed subgrid-scale (SGS) models based on high-pass filter (HPF), mean-field eddy viscosity (MFEV), and Smagorinsky (SMG) with no-slip and traction boundary conditions, provided with low-order statistics, convergence and turbulent structure analysis. The model fidelity and scaling performance of Nek5000/RS on DOE’s leadership computing platforms in comparison to those of AMR-Wind, a block-structured second-order finite-volume code with adaptive-mesh-refinement capabilities, are discussed in [3].

17 WIND ENERGY↗

Advances in Multi-Physics Modeling for Parallel Finite Element Code Suite ACE3P

SLAC developed ACE3P code suite is a comprehensive set of parallel finite-element codes for multi-physics modeling of particle accelerators. Running on massively parallel computer platforms for high fidelity and high accuracy simulation, ACE3P enables rapid virtual prototyping of accelerator and RF component design, optimization and analysis. Furthermore, recent advances of ACE3P have been focused on the implementation of advanced numerical algorithms, enhancement of multi-physics capabilities, and improvement of code performance on state-of-the-art high performance computing (HPC) platforms for large-scale RF modeling in accelerator applications and beyond. In this paper, we will present a nonlinear electromagnetic (EM) eigensolver for damping calculation of resonant modes for cavities coupled with waveguides, the thermal solver for accurate evaluation of heat loads in superconducting RF (SRF) cavity power coupler, the elastic solver for investigation of SRF cavity cryomodule (CM) bowing due to thermal effect, the frequency domain mechanical solver for studying SRF cavity mechanical oscillation modes, the particle tracking code for dark current simulation in SRF CM using a hybrid MPI+OpenMP parallel programing model, the time domain solver for high fidelity modeling of EM wave propagation for medical device in human body.

97 MATHEMATICS AND COMPUTING↗

Implementing Molecular Dynamics for Hybrid High Performance Computers - 1. Short Range Forces

The use of accelerators such as general-purpose graphics processing units (GPGPUs) have become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power requirements. Hybrid high performance computers, machines with more than one type of floating-point processor, are now becoming more prevalent due to these advantages. In this work, we discuss several important issues in porting a large molecular dynamics code for use on parallel hybrid machines - 1) choosing a hybrid parallel decomposition that works on central processing units (CPUs) with distributed memory and accelerator cores with shared memory, 2) minimizing the amount of code that must be ported for efficient acceleration, 3) utilizing the available processing power from both many-core CPUs and accelerators, and 4) choosing a programming model for acceleration. We present our solution to each of these issues for short-range force calculation in the molecular dynamics package LAMMPS. We describe algorithms for efficient short range force calculation on hybrid high performance machines. We describe a new approach for dynamic load balancing of work between CPU and accelerator cores. We describe the Geryon library that allows a single code to compile with both CUDA and OpenCL for use on a variety of accelerators. Finally, we present results on a parallel test cluster containing 32 Fermi GPGPUs and 180 CPU cores.

43 PARTICLE ACCELERATORS↗

Radiation shielding for the Fermilab Vertical Cavity Test Facility

The results of radiation shielding studies for the vertical test cryostat VTS1 at Fermilab performed with the codes FISHPACT and MARS15 are presented and discussed. The analysis is focused on operations with two RF cavities in the cryostat. The vertical cavity test facility (VCTF) for superconducting RF cavities in Industrial Building 1 at Fermilab has been in operation since 2007. The facility currently consists of a single vertical test cryostat VTS1. Radiation shielding for VTS1 was designed for operations with single 9-cell 1.3 GHz cavities, and the shielding calculations were performed using a simplified model of field emission as the radiation source. The operations are proposed to be extended in such a way that two RF cavities will be in VTS1 at a time, one above the other, with tests for each cavity performed sequentially. In such a case the radiation emitted during the tests from the lower cavity can, in part, bypass the initially designed shielding which can lead to a higher dose in the building. Space for additional shielding, either internal or external to VTS1, is limited. Therefore, a re-evaluation of the radiation shielding was performed. An essential part of the present analysis is in using realistic models for cavity geometry and spatial, angular and energy distributions of field-emitted electrons inside the cavities. The calculations were performed with the computer codes FISHPACT and MARS15.

43 PARTICLE ACCELERATORS↗

Universality of the acceleration due to gravity on the surface of a rapidly rotating neutron star

On the surface of a rapidly rotating neutron star, the effective centrifugal force decreases the effective acceleration due to gravity (as measured in the rotating frame) at the equator while increasing the acceleration at the poles due to the centrifugal flattening of the star into an oblate spheroid. We compute the effective gravitational acceleration for relativistic rapidly rotating neutron stars and show that for a star with mass M, equatorial radius R{sub e} , and angular velocity Ω, the deviations of the effective acceleration due to gravity from the nonrotating case take on a universal form that depends only on the compactness ratio M/R{sub e} , the dimensionless square of the angular velocity Ω{sup 2}R{sub e}{sup 3}/GM, and the latitude on the star's surface. This dependence is universal, in that it has very little dependence on the neutron star's equation of state. The effective gravity is expanded in the slow-rotation limit to show the dependence on the effective centrifugal force, oblate shape of the star, and the quadrupole moment of the gravitational field. In addition, an empirical fit and simple formula for the effective gravity is found. We find that the increase in the acceleration due to gravity at the poles is of the same order of magnitude as the decrease in the effective acceleration due to gravity at the equator for all realistic value of mass, radius, and spin. For neutron stars that spin with frequencies near 600 Hz, the difference between the effective gravity at the poles and the equator is about 20%.

79 ASTRONOMY AND ASTROPHYSICS↗

Language models for the prediction of SARS-CoV-2 inhibitors

The COVID-19 pandemic highlights the need for computational tools to automate and accelerate drug design for novel protein targets. We leverage deep learning language models to generate and score drug candidates based on predicted protein binding affinity. We pre-trained a deep learning language model (BERT) on ∼9.6 billion molecules and achieved peak performance of 603 petaflops in mixed precision. Our work reduces pre-training time from days to hours, compared to previous efforts with this architecture, while also increasing the dataset size by nearly an order of magnitude. For scoring, we fine-tuned the language model using an assembled set of thousands of protein targets with binding affinity data and searched for inhibitors of specific protein targets, SARS-CoV-2 Mpro and PLpro. We utilized a genetic algorithm approach for finding optimal candidates using the generation and scoring capabilities of the language model. Our generalizable models accelerate the identification of inhibitors for emerging therapeutic targets.

Blanchard, Andrew E.↗

Solving large-scale sparse eigenvalue problems and linear systems of equations for accelerator modeling

The solutions of sparse eigenvalue problems and linear systems constitute one of the key computational kernels in the discretization of partial differential equations for the modeling of linear accelerators. The computational challenges faced by existing techniques for solving those sparse eigenvalue problems and linear systems call for continuing research to improve on the algorithms so that ever increasing problem size as required by the physics application can be tackled. Under the support of this award, the filter algorithm for solving large sparse eigenvalue problems was developed at Stanford to address the computational difficulties in the previous methods with the goal to enable accelerator simulations on then the world largest unclassified supercomputer at NERSC for this class of problems. Specifically, a new method, the Hemitian skew-Hemitian splitting method, was proposed and researched as an improved method for solving linear systems with non-Hermitian positive definite and semidefinite matrices.

43 PARTICLE ACCELERATORS↗

XaaS: Acceleration as a Service to Enable Productive High-Performance Cloud Computing

High-performance computing (HPC) and the cloud have evolved independently, specializing their innovations into performance or productivity. Acceleration as a Service (XaaS) is a recipe to empower both fields with a shared execution platform that provides transparent access to computing resources, regardless of the underlying cloud or HPC service provider. Bridging HPC and cloud advancements, XaaS presents a unified architecture built on performance-portable containers. Here, our converged model concentrates on low-overhead, high-performance communication and computing, targeting resource-intensive workloads from climate simulations to machine learning. XaaS lifts the restricted allocation model of Function as a Service (FaaS), allowing users to benefit from the flexibility and efficient resource utilization of serverless computing while supporting long-running and performance-sensitive workloads from HPC.

97 MATHEMATICS AND COMPUTING↗

Introduction to CERN

Summer Student Lecture Programme Introduction. The mission of CERN; push back the frontiers of knowledge, e.g. the secrets of the Big Bang...what was the matter like within the first moments of the Universe's existence? You have to develop new technologies for accelerators and detectors (also information technology--the Web and the GRID and medicine--diagnosis and therapy). There are three key technology areas at CERN; accelerating, particle detection, large-scale computing.

43 PARTICLE ACCELERATORS↗