Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallelization efficiency”

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

Highly scalable distributed-memory sparse triangular solution algorithms.

This paper presents a highly efficient distributed-memory parallel sparse triangular solver. The triangular solution phase is often performed following factorization phase in the sparse linear solvers and has become increasingly computa­tionally expensive for direct solvers with many right hand sides (RHSs) or preconditioned iterative solvers. However, the low arithmetic intensity and sequential nature of the triangular solve algorithm pose performance challenges for its large-scale distributed-memory parallelization. In this work, we propose several strategies to enhance scalability of an algorithm with 2D block cyclic process layout. First, an asynchronous binary-tree-based communication scheme implemented via non-blocking MPI functions is leveraged to broadcast partial solutions and reduce partial updates among a subset of processes for each block column and row of the triangular matrix, respectively. This scheme reduces message latency, improves communication load balance and significantly accelerates asynchronous execution of the triangular solve. In addition, efficient BLAS operations and threading implementations are exploited to accelerate local computations and further reduce process idle time. The proposed strategies are implemented in SuperLU_DIST and numerical experiments show up to 4.4x improvement with one right-hand side and up to 6.1x improvement with 50 right- hand sides on 4096 processes, compared to the current release. This is the first time that sparse triangular solution is demonstrated strong scaling on more than 4000 cores.

Liu, Yang↗
DOE OSTI2021

Parallel physics-informed neural networks via domain decomposition

Here we develop a distributed framework for the physics-informed neural networks (PINNs) based on two recent extensions, namely conservative PINNs (cPINNs) and extended PINNs (XPINNs), which employ domain decomposition in space and in time-space, respectively. This domain decomposition endows cPINNs and XPINNs with several advantages over the vanilla PINNs, such as parallelization capacity, large representation capacity, efficient hyperparameter tuning, and is particularly effective for multi-scale and multi-physics problems. Here, we present a parallel algorithm for cPINNs and XPINNs constructed with a hybrid programming model described by MPI + X, where X ∈ {CPUs, GPUs}. The main advantage of cPINN and XPINN over the more classical data and model parallel approaches is the flexibility of optimizing all hyperparameters of each neural network separately in each subdomain. We compare the performance of distributed cPINNs and XPINNs for various forward problems, using both weak and strong scalings. Our results indicate that for space domain decomposition, cPINNs are more efficient in terms of communication cost but XPINNs provide greater flexibility as they can also handle time-domain decomposition for any differential equations, and can deal with any arbitrarily shaped complex subdomains. To this end, we also present an application of the parallel XPINN method for solving an inverse diffusion problem with variable conductivity on the United States map, using ten regions as subdomains.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

Addressing Load Imbalance in Bioinformatics and Biomedical Applications: Efficient Scheduling across Multiple GPUs

Computational bioinformatics and biomedical applications frequently contain heterogeneously sized units of work or tasks, for instance due to variability in the sizes of biological sequences and molecules. Variable-sized workloads lead to load imbalances in parallel implementations which detract from efficiency and performance. Many modern computing resources now have multiple graphics processing units(GPUs) per computer for acceleration. These multiple GPU resources need to be used efficiently through balancing of workloads across the GPUs. OpenMP is a portable directive-based parallel programming API used ubiquitously in bioscience applications to program CPUs; recently, the use of OpenMP directives for GPU acceleration has become possible. Here, motivated by experiences with imbalanced loads in GPU-accelerated bioinformatics applications, we address the load balancing problem using OpenMP task-to-GPU scheduling combined with OpenMP GPU offloading for multiply heterogeneous workloads – loads with both variable input sizes, and simultaneously, variable convergence rates for algorithms with a stochastic component – scheduled across multiple GPUs. We aim to develop strategies which are both easy to use and have lower overheads, and may be incorporated incrementally in existing programs which already make use of OpenMP for CPU-based threading in order to make use of multi-GPU computers. We test different combinations of input size variability and convergence rate variability, and characterize the effects of these different scenarios on the performance of scheduling strategies across multiple GPUs with OpenMP. We present several dynamic scheduling solutions for different parallel patterns, explore optimizations, and provide publicly available example computational kernels to make these strategies easy to use in programs. This work will enable application developers to efficiently and easily use multiple GPUs for imbalanced workloads found in bioinformatics and biomedical applications.

Thavappiragasam, Mathialakan↗
DOE OSTI2021

Energy-efficient Mott activation neuron for full-hardware implementation of neural networks

To circumvent the von Neumann bottleneck, substantial progress has been made towards in-memory computing with synaptic devices. However, compact nanodevices implementing non-linear activation functions are required for efficient full-hardware implementation of deep neural networks. Here, in this work, we present an energy-efficient and compact Mott activation neuron based on vanadium dioxide and its successful integration with a conductive bridge random access memory (CBRAM) crossbar array in hardware. The Mott activation neuron implements the rectified linear unit function in the analogue domain. The neuron devices consume substantially less energy and occupy two orders of magnitude smaller area than those of analogue complementary metal–oxide semiconductor implementations. The LeNet-5 network with Mott activation neurons achieves 98.38% accuracy on the MNIST dataset, close to the ideal software accuracy. We perform large-scale image edge detection using the Mott activation neurons integrated with a CBRAM crossbar array. Our findings provide a solution towards large-scale, highly parallel and energy-efficient in-memory computing systems for neural networks.

electrical and electronic engineering↗
DOE OSTI2019

Parallel programming of an ionic floating-gate memory array for scalable neuromorphic computing

Neuromorphic computers could overcome efficiency bottlenecks inherent to conventional computing through parallel programming and readout of artificial neural network weights in a crossbar memory array. However, selective and linear weight updates and <10-nanoampere read currents are required for learning that surpasses conventional computing efficiency. We introduce an ionic floating-gate memory array based on a polymer redox transistor connected to a conductive-bridge memory (CBM). Selective and linear programming of a redox transistor array is executed in parallel by overcoming the bridging threshold voltage of the CBMs. Synaptic weight readout with currents <10 nanoamperes is achieved by diluting the conductive polymer with an insulator to decrease the conductance. The redox transistors endure >1 billion write-read operations and support >1-megahertz write-read frequencies.

bio-inspired, energy storage (including batteries ↗
DOE OSTI2020

Extended Physics-Informed Neural Networks (XPINNs): A Generalized Space-Time Domain Decomposition Based Deep Learning Framework for Nonlinear Partial Differential Equations

Here we propose a generalized space-time domain decomposition approach for the physics-informed neural networks (PINNs) to solve nonlinear partial differential equations (PDEs) on arbitrary complex-geometry domains. The proposed framework, named eXtended PINNs ( X P I N N s ), further pushes the boundaries of both PINNs as well as conservative PINNs (cPINNs), which is a recently proposed domain decomposition approach in the PINN framework tailored to conservation laws. Compared to PINN, the XPINN method has large representation and parallelization capacity due to the inherent property of deployment of multiple neural networks in the smaller subdomains. Unlike cPINN, XPINN can be extended to any type of PDEs. Moreover, the domain can be decomposed in any arbitrary way (in space and time), which is not possible in cPINN. Thus, XPINN offers both space and time parallelization, thereby reducing the training cost more effectively. In each subdomain, a separate neural network is employed with optimally selected hyperparameters, e.g., depth/width of the network, number and location of residual points, activation function, optimization method, etc. A deep network can be employed in a subdomain with complex solution, whereas a shallow neural network can be used in a subdomain with relatively simple and smooth solutions. We demonstrate the versatility of XPINN by solving both forward and inverse PDE problems, ranging from one-dimensional to three-dimensional problems, from time-dependent to time-independent problems, and from continuous to discontinuous problems, which clearly shows that the XPINN method is promising in many practical problems. The proposed XPINN method is the generalization of PINN and cPINN methods, both in terms of applicability as well as domain decomposition approach, which efficiently lends itself to parallelized computation. The XPINN code is available on h t t p s : / / g i t h u b . c o m / A m e y a J a g t a p / X P I N N s .

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

High-Level Synthesis of Parallel Specifications Coupling Static and Dynamic Controllers

The increased need for efficient ways to implement domain-specific accelerators is driving design methodologies towards the use of abstractions higher than the Register Transfer Level (RTL). In this scenario, High Level Synthesis (HLS) plays a significant role by enabling the automatic generation of custom hardware accelerators starting from high level descriptions (e.g., C code). Conventional HLS tools exploit parallelism mostly at the Instruction Level (ILP). They statically schedule the input specifications, and build centralized Finite State Machine (FSM) controllers. However, aggressive exploitation of ILP in many applications has diminishing returns and, usually, centralized approaches do not efficiently exploit coarser parallelism because FSMs are inherently serial. In this paper we present a HLS framework able to synthesize applications that, beside ILP, also expose Task Level Parallelism (TLP). An application can expose TLP through annotations that identify the parallel functions (i.e., tasks). To generate accelerators that efficiently execute concur- rent tasks, we need to solve several issues: devise a mechanism to support concurrent execution flows, exploit memory parallelism, and manage synchronization. To support concurrent execution flows, we introduce a novel adaptive controller. The adaptive controller is composed of a set of interacting control elements that independently manage the execution of a single operation or function call. These control elements check dependencies and resource constraints at runtime, enabling as soon as possible execution. To support parallel access to shared memories and synchronization, we introduce a novel Hierarchical Memory Interface (HMI). With respect to previous solutions, the proposed interface supports multi-ported memories and atomic memory operations, which commonly occur in parallel programming. Our framework can generate the hardware implementation of C functions by employing two different approaches, depending on its characteristics. If a function exposes TLP, then the framework generates hardware implementations based on the adaptive controller. Otherwise, the framework implements the function by exploiting a more conventional FSM approach, which is optimized for ILP exploitation. We evaluate our framework on a set of parallel applications, and show substantial performance improvements (average speedup of 4.7) with limited area over- heads (average area increase of 5.48 times).

Castellana, Vito G.↗
DOE OSTI2021

High-Level Synthesis of Parallel Specifications Coupling Static and Dynamic Controllers

The increased need for efficient ways to implement domain-specific accelerators is driving design methodologies towards the use of abstractions higher than the Register Transfer Level (RTL). In this scenario, High Level Synthesis (HLS) plays a significant role by enabling the automatic generation of custom hardware accelerators starting from high level descriptions (e.g., C code). Conventional HLS tools exploit parallelism mostly at the Instruction Level (ILP). They statically schedule the input specifications, and build centralized Finite State Machine (FSM) controllers. However, aggressive exploitation of ILP in many applications has diminishing returns and, usually, centralized approaches do not efficiently exploit coarser parallelism because FSMs are inherently serial. In this paper we present a HLS framework able to synthesize applications that, beside ILP, also expose Task Level Parallelism (TLP). An application can expose TLP through annotations that identify the parallel functions (i.e., tasks). To generate accelerators that efficiently execute concur- rent tasks, we need to solve several issues: devise a mechanism to support concurrent execution flows, exploit memory parallelism, and manage synchronization. To support concurrent execution flows, we introduce a novel adaptive controller. The adaptive controller is composed of a set of interacting control elements that independently manage the execution of a single operation or function call. These control elements check dependencies and resource constraints at runtime, enabling as soon as possible execution. To support parallel access to shared memories and synchronization, we introduce a novel Hierarchical Memory Interface (HMI). With respect to previous solutions, the proposed interface supports multi-ported memories and atomic memory operations, which commonly occur in parallel programming. Our framework can generate the hardware implementation of C functions by employing two different approaches, depending on its characteristics. If a function exposes TLP, then the framework generates hardware implementations based on the adaptive controller. Otherwise, the framework implements the function by exploiting a more conventional FSM approach, which is optimized for ILP exploitation. We evaluate our framework on a set of parallel applications, and show substantial performance improvements (average speedup of 4.7) with limited area over- heads (average area increase of 5.48 times).

Castellana, Vito G.↗
DOE OSTI2011

Effect of electron flow on the ordinary-extraordinary mode conversion

Ordinary-extraordinary mode conversion in the electron cyclotron frequency range is revisited in the presence of a flowing electron component. The analytical expressions of optimal parallel refraction index and conversion efficiency are obtained from a one-dimensional cold plasma model. The presence of flowing electrons leads to an outward shift of the conversion layer and therefore increases the optimal value of parallel refraction index. If this effect is not considered, the efficiency of mode conversion degenerates. In typical tokamak plasmas, this degeneration is about a few percentages, which may induce the reflection of several tens of kilowatts of power from the cutoff layer when injecting megawatts of ECRF power into fusion plasma.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2014

Reliable and Efficient Distributed Checkpointing System for Grid Environments

In Fine-Grained Cycle Sharing (FGCS) systems, machine owners voluntarily share their unused CPU cycles with guest jobs, as long as their performance degradation is tolerable. However, unpredictable evictions of guest jobs lead to fluctuating completion times. Checkpoint-recovery is an attractive mechanism for recovering from such “failures”. Today’s FGCS systems often use expensive, high-performance dedicated checkpoint servers. However, in geographically distributed clusters, this may incur high checkpoint transfer latencies. Here we present a distributed checkpointing system called FALCON that uses available disk resources of the FGCS machines as shared checkpoint repositories. However, an unavailable storage host may lead to loss of checkpoint data. Therefore, we model the failures of a storage host and develop a prediction algorithm for choosing reliable checkpoint repositories. We experiment with FALCON in the university-wide Condor testbed at Purdue and show improved and consistent performance for guest jobs in the presence of irregular resource availability.

checkpoint↗
DOE OSTI2020

Impact of moderate pump–Stokes chirp on femtosecond coherent anti-Stokes Raman scattering spectra

The effects of inducing moderate chirp in the pump and Stokes beams in chirped-probe-pulse femtosecond coherent anti-Stokes Raman scattering (CARS) spectroscopy are discussed. This is an important issue for measurements in high-pressure systems where the CARS beams must be transmitted through glass of significant thickness to reach the probe volume of interest. The effects were investigated both experimentally, by inserting disks of SF-11 glass into the pump and Stokes beam paths, and theoretically, by incorporating pulse chirp into our time-dependent density matrix (TDDM) simulations. Experimentally, we found that inducing moderate pump and Stokes chirp produced significant spectral narrowing of the nonresonant four-wave mixing contribution to the CARS signal, and this allowed us to control the resonant excitation bandwidth to eliminate interferences from neighboring species. The temperature measurement accuracy and precision were essentially unchanged with respect to the unchirped pump and Stokes case. The effect of moderate pump and Stokes chirp on the narrowing of the Raman excitation efficiency envelope was investigated by solving the full set of TDDM through direct numerical integration. The numerical efficiency of the solutions was enhanced dramatically by developing a parallel version of the TDDM code. The calculated Raman excitation efficiency profile was incorporated in our phenomenological spectral fitting codes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2015

Jagged Tiling for Intra-tile Parallelism and Fine-Grain Multithreading

In this paper, we have developed a novel methodology that takes into consideration multithreaded many-core designs to better utilize memory/processing resources and improve memory residence on tileable applications. It takes advantage of polyhedral analysis and transformation in the form of PLUTO, combined with a highly optimized finegrain tile runtime to exploit parallelism at all levels. The main contributions of this paper include the introduction of multi-hierarchical tiling techniques that increases intra tile parallelism; and a data-flow inspired runtime library that allows the expression of parallel tiles with an efficient synchronization registry. Our current implementation shows performance improvements on an Intel Xeon Phi board up to 32.25% against instances produced by state-of-the-art compiler frameworks for selected stencil applications.

compiler↗
DOE OSTI2024

Efficient distributed continual learning for steering experiments in real-time

Deep learning has emerged as a powerful method for extracting valuable information from large volumes of data. However, when new training data arrives continuously (i.e., is not fully available from the beginning), incremental training suffers from catastrophic forgetting (i.e., new patterns are reinforced at the expense of previously acquired knowledge). Training from scratch each time new training data becomes available would result in extremely long training times and massive data accumulation. Rehearsal-based continual learning has shown promise for addressing the catastrophic forgetting challenge, but research to date has not addressed performance and scalability. To fill this gap, we propose an approach based on a distributed rehearsal buffer that efficiently complements data-parallel training on multiple GPUs to achieve high accuracy, short runtime, and scalability. It leverages a set of buffers (local to each GPU) and uses several asynchronous techniques for updating these local buffers in an embarrassingly parallel fashion, all while handling the communication overheads necessary to augment input minibatches using unbiased, global sampling. We further propose a generalization of rehearsal buffers to support both classification and generative learning tasks, as well as more advanced rehearsal strategies (notably Dark Experience Replay, leveraging knowledge distillation). We illustrate this approach with a real-life HPC streaming application from the domain of ptychographic image reconstruction. Furthermore, we run extensive experiments on up to 128 GPUs of the ThetaGPU supercomputer to compare our approach with baselines representative of training-from-scratch (the upper bound in terms of accuracy) and incremental training (the lower bound). Results show that rehearsal-based continual learning achieves a top-5 validation accuracy close to the upper bound, while simultaneously exhibiting a runtime close to the lower bound.

Asynchronous data management↗
DOE OSTI2018

A purely Lagrangian method for simulating the shallow water equations on a sphere using smooth particle hydrodynamics

It has long been suggested that a purely Lagrangian solution to global-scale atmospheric/oceanic flows can potentially outperform tradition Eulerian schemes. Meanwhile, a demonstration of a scalable and practical framework remains elusive. Motivated by recent progress in particle-based methods when applied to convection dominated flows, this work presents a fully Lagrangian method for solving the inviscid shallow water equations on a rotating sphere in a smooth particle hydrodynamics framework. To avoid singularities at the poles, the governing equations are solved in Cartesian coordinates, augmented with a Lagrange multiplier to ensure that fluid particles are constrained to the surface of the sphere. An underlying grid in spherical coordinates is used to facilitate efficient neighbor detection and parallelization. The method is applied to a suite of canonical test cases, and conservation, accuracy, and parallel performance are assessed..

Computer Science↗
DOE OSTI2018

Calibrated simulation analysis for integration of evaporative cooling and radiant cooling system for different Indian climatic zones

Radiant cooling system (RCS) has proven to be an energy efficient system for meeting a building's cooling requirements. RCS is energy efficient and provides better thermal comfort compared to conventional all-air heating ventilation air conditioning (HVAC) system. To further improve the efficiency of RCS, a parallel evaporative cooling system (cooling tower) is coupled with RCS and analyzed for different climatic condition using calibrated model. A statistical analysis of weather files (based on wet bulb temperature), was used to identify the availability of useful water for cooling tower integration with RCS. A comprehensive simulation feasibility study of the application of cooling tower in RCS was performed for different cities to cover every climatic zones of India. It was found that in summer, the wet bulb temperature (WBT) of different climatic zones, except warm-humid, is suitable for the integration of cooling tower with RCS. An experimental setup was designed and developed for integration of parallel chiller and cooling tower with panel-based RCS for cooling and a dedicated outdoor air system (DOAS) for dehumidification and ventilation. Experiments were conducted for chiller and cooling tower operated RCS in Composite climate of Jaipur, Rajasthan, India. Building Energy Models (BEM) were developed for both the chiller-operated RCS and cooling tower-operated RCS in EnergyPlus and calibrated with the measured data. Using the calibrated models, performance of the system was analyzed for different climatic zones of India. A chiller-operated RCS was considered as a baseline to compare the annual energy saving potential and monthly performance of the cooling tower integrated RCS. Here, in the cooling tower–operated RCS, a total annual savings of 7% in hot and dry climates, 11% in composite climates and 20% in temperate climates was achieved compared to the chiller-operated RCS.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Efficient Treatment of Large Active Spaces through Multi-GPU Parallel Implementation of Direct Configuration Interaction

In this study, we have extended our graphical processing unit (GPU)-accelerated direct configuration interaction program to multiple devices, reducing iteration times for configuration spaces of 165 million determinants to only 3 s using NVIDIA P100 GPUs. Similar improvements in the one- and two-particle reduced density matrix formation allow for fast analytical energy gradients and electronic properties. Our parallel algorithm enables the calculation of arbitrarily large configuration spaces (limited only by available system memory), with iteration times of 13 min for an active space of 18 electrons in 18 orbitals (2.4 billion determinants) using six consumer grade NVIDIA 1080Ti GPUs. These advances enable routine molecular dynamics simulations, geometry optimizations, and absorption spectrum calculations for molecules with large configuration spaces, a task that has heretofore required massive computational effort. In this work, we demonstrate the utility of our program by generating the absorption spectrum for diphenyl acetylene at the floating occupation molecular orbital complete active space configuration interaction level of theory. Lastly, several active spaces were investigated to assess the dependence of spectral features on orbital space dimension.

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

Automated “Cells-To-Peptides” Sample Preparation Workflow for High-Throughput, Quantitative Proteomic Assays of Microbes

Mass spectrometry-based quantitative proteomic analysis has proven valuable for clinical and biotechnology-related research and development. Improvements in sensitivity, resolution, and robustness of mass analyzers have also added value. However, manual sample preparation protocols are often a bottleneck for sample throughput and can lead to poor reproducibility, especially for applications where thousands of samples per month must be analyzed. To alleviate these issues, we developed a “cells-to-peptides” automated workflow for Gram-negative bacteria and fungi that includes cell lysis, protein precipitation, resuspension, quantification, normalization, and tryptic digestion. The workflow takes 2 h to process 96 samples from cell pellets to the initiation of the tryptic digestion step and can process 384 samples in parallel. We measured the efficiency of protein extraction from various amounts of cell biomass and optimized the process for standard liquid chromatography–mass spectrometry systems. The automated workflow was tested by preparing 96 Escherichia coli samples and quantifying over 600 peptides that resulted in a median coefficient of variation of 15.8%. Similar technical variance was observed for three other organisms as measured by highly multiplexed LC-MRM–MS acquisition methods. These results show that this automated sample preparation workflow provides robust, reproducible proteomic samples for high-throughput applications.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Solving sparse finite element problems on neuromorphic hardware

The finite element method (FEM) is one of the most important and ubiquitous numerical methods for solving partial differential equations (PDEs) on computers for scientific and engineering discovery. Applying the FEM to larger and more detailed scientific models has driven advances in high-performance computing for decades. Here we demonstrate that scalable spiking neuromorphic hardware can directly implement the FEM by constructing a spiking neural network that solves the large, sparse, linear systems of equations at the core of the FEM. We show that for the Poisson equation, a fundamental PDE in science and engineering, our neural circuit achieves meaningful levels of numerical accuracy and close to ideal scaling on modern, inherently parallel and energy-efficient neuromorphic hardware, specifically Intel’s Loihi 2 neuromorphic platform. We illustrate extensions to irregular mesh geometries in both two and three dimensions as well as other PDEs such as linear elasticity. Our spiking neural network is constructed from a recurrent network model of the brain’s motor cortex and, in contrast to black-box deep artificial neural network-based methods for PDEs, directly translates the well-understood and trusted mathematics of the FEM to a natively spiking neuromorphic algorithm.

Applied mathematics↗