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 1,441 records · Page 80

Enhancing ChatPORT with CUDA-to-SYCL Kernel Translation Capability

Large Language Models (LLMs) have shown strong capabilities in general code translation. However, code translation involving parallel programming models remains largely unexplored. This work enhances the capabilities of code LLMs in CUDA-to-SYCL kernel translation with parameter-efficient fine-tuning. The resultant fine-tuned LLM, called ChatPORT, is an effort to provide high-fidelity translations from one programming model to another. We describe the preparation of datasets from heterogeneous computing benchmarks for model fine-tuning and testing, the parameter-efficient fine-tuning of 19 open-source code models ranging in size from 0.5 to 34 billion parameters and evaluate the correctness rates of the SYCL kernels by the fine-tuned models. The experimental results show that most code models fail to translate CUDA codes to SYCL correctly. However, fine-tuning these models using a small set of CUDA and SYCL kernels can enhance the capabilities of these models in kernel translation. Depending on the sizes of the models, the correctness rate ranges from 19.9% to 81.7% for a test dataset of 62 CUDA kernels.

Jin, Zheming [ORNL] (ORCID:000000027197780X)↗

CICE, The Los Alamos Sea Ice Model

The Los Alamos sea ice model (CICE) is the result of an effort to develop a computationally efficient sea ice component for a fully coupled atmosphere–land–ocean–ice global climate model. It was originally designed to be compatible with the Parallel Ocean Program (POP), an ocean circulation model developed at Los Alamos National Laboratory for use on massively parallel computers. CICE has several interacting components: a vertical thermodynamic model that computes local growth rates of snow and ice due to vertical conductive, radiative and turbulent fluxes, along with snowfall; an elastic-viscous-plastic model of ice dynamics, which predicts the velocity field of the ice pack based on a model of the material strength of the ice; an incremental remapping transport model that describes horizontal advection of the areal concentration, ice and snow volume and other state variables; and a ridging parameterization that transfers ice among thickness categories based on energetic balances and rates of strain. It also includes a biogeochemical model that describes evolution of the ice ecosystem. The CICE sea ice model is used for climate research as one component of complex global earth system models that include atmosphere, land, ocean and biogeochemistry components. It is also used for operational sea ice forecasting in the polar regions and in numerical weather prediction models.

Hunke, Elizabeth↗

smoothG

smoothG is a collection of parallel C++ classes/functions that algebraically constructs reduced models of different resolutions from a given high-fidelity graph model. In addition, smoothG also provides efficient linear solvers for the reduced models. Other than pure graph problem, the software finds its application in subsurface flow and power grid simulations in which graph Laplacians are found

Barker, AndrewT.↗

Rendering and Compositing Infrastructure Improvements to VisIt for Insitu Rendering

Compared to posthoc rendering, insitu rendering often generates larger numbers of images, as a result rendering performance and scalability are critical in the insitu setting. In this work we present improvements to VisIt's rendering and compositing infrastructure that deliver increased performance and scalability in both posthoc and insitu settings. We added the capability for alpha blend compositing and use it with ordered compositing when datasets have disjoint block domain decomposition to optimize the rendering of transparent geometry. We also made improvements that increase overall efficiency by reducing communication and data movement and have addressed a number of performance issues. We structured our code to take advantage of SIMD parallelization and use threads to overlap communication and compositing. We tested our improvements on a 20 core workstation using 8 cores to render geometry generated from a $256^3$ cosmology dataset and on a Cray XC31 using 512 cores to render geometry generated from a $2000^2 \times 800$ plasma dataset. Our results show that ordered compositing provides a speed up of up to $4 \times$ over the current sort first strategy. The other improvements resulted in modest speed up with one notable exception where we achieve up to $40 \times$ speed up of rendering and compositing of opaque geometry when both opaque and transparent geometry are rendered together. We also investigated the use of depth peeling, but found that the implementation provided by VTK is substantially slower,both with and without GPU acceleration, than a local camera order sort.

97 MATHEMATICS AND COMPUTING↗

Community Project for Accelerator Science and Simulation (ComPASS)

After concluding our initial exercise (solving a simplified statistical inverse problem with unknown parameter laser intensity) of coupling Vorpal and our parallel statistical library QUESO, we shifted the application focus to DLA. Our efforts focused on developing a Gaussian process (GP) emulator within QUESO for efficient optimization of power couplers within woodpiles. The smaller simulation size (compared with LPA) allows for sufficient “training runs” to develop a reasonable GP statistical emulator for a parameter space of moderate dimension.

43 PARTICLE ACCELERATORS↗

Seamless Hybrid-integrated Interconnect NEtwork (SHINE)

Large-scale date centers are becoming increasingly limited by the capacity of interconnects due to the rapid increasing power consumption and bandwidth demands driven by the rigorous down scaling of CMOS critical dimensions in according with the Moore’s Law. As a result of the continuing bandwidth scaling, optical interconnects based on parallel optics, where data are simultaneously transmitted along more than one lane, have becoming a preferred interconnected solution over copper cables. To building an efficient optical interconnect system, optical coupling between different elements, such as optical fibers, photonics chips, is a critical step, which is challenging due to the mode mismatch of varies optical ports. In this project, we proposed and demonstrated an optical interfacing scheme, which provided a light coupling approach with low energy loss, broadband operation, high alignment tolerance and compatible with scalable fabrication and testing. Here, the mode transformation is achieved by a total internal reflection from a chip integrated microscale freefrom surface. Due to the enormous light wavefront manipulating ability associated with the gigantic design space, the freeform couplers platform is universally applicable to a wide range of optical coupling scenarios, such as the interfacing between a waveguide to a fiber, another waveguide, surface normal device, or free space.

42 ENGINEERING↗

Scalable Second Order Optimization for Machine Learning

Many machine learning (ML) training tasks are essentially optimization processes that would at first glance appear eminently parallelizable and scalable. However, effective acceleration of these tasks with scalable parallel hardware has proven to be elusive. While standard methods for machine learning, e.g., stochastic gradient descent (SGD) for DNNs, tend to be resource efficient, they appear to be fundamentally sequential in nature.

97 MATHEMATICS AND COMPUTING↗

Electrochemical cell stack assembly

Multiple stacks of tubular electrochemical cells having a dense electrolyte disposed between an anode and a cathode preferably deposited as thin films arranged in parallel on stamped conductive interconnect sheets or ferrules. The stack allows one or more electrochemical cell to malfunction without disabling the entire stack. Stack efficiency is enhanced through simplified gas manifolding, gas recycling, reduced operating temperature and improved heat distribution.

30 DIRECT ENERGY CONVERSION↗

CICE, The Los Alamos Sea Ice Model

The Los Alamos sea ice model (CICE) is the result of an effort to develop a computationally efficient sea ice component for a fully coupled atmosphere–land–ocean–ice global climate model. It was originally designed to be compatible with the Parallel Ocean Program (POP), an ocean circulation model developed at Los Alamos National Laboratory for use on massively parallel computers. CICE has several interacting components: a vertical thermodynamic model that computes local growth rates of snow and ice due to vertical conductive, radiative and turbulent fluxes, along with snowfall; an elastic-viscous-plastic model of ice dynamics, which predicts the velocity field of the ice pack based on a model of the material strength of the ice; an incremental remapping transport model that describes horizontal advection of the areal concentration, ice and snow volume and other state variables; and a ridging parameterization that transfers ice among thickness categories based on energetic balances and rates of strain. It also includes a biogeochemical model that describes evolution of the ice ecosystem. The CICE sea ice model is used for climate research as one component of complex global earth system models that include atmosphere, land, ocean and biogeochemistry components. It is also used for operational sea ice forecasting in the polar regions and in numerical weather prediction models.

Hunke, Elizabeth↗

An Asynchronous Task-based Fan-Both Sparse Cholesky Solver

Systems of linear equations arise at the heart of many scientific and engineering applications. Many of these linear systems are sparse; i.e., most of the elements in the coefficient matrix are zero. Direct methods based on matrix factorizations are sometimes needed to ensure accurate solutions. For example, accurate solution of sparse linear systems is needed in shift-invert Lanczos to compute interior eigenvalues. The performance and resource usage of sparse matrix factorizations are critical to time-to-solution and maximum problem size solvable on a given platform. In many applications, the coefficient matrices are symmetric, and exploiting symmetry will reduce both the amount of work and storage cost required for factorization. When the factorization is performed on large-scale distributed memory platforms, communication cost is critical to the performance of the algorithm. At the same time, network topologies have become increasingly complex, so that modern platforms exhibit a high level of performance variability. This makes scheduling of computations an intricate and performance-critical task. In this paper, we investigate the use of an asynchronous task paradigm, one-sided communication and dynamic scheduling in implementing sparse Cholesky factorization (symPACK) on large-scale distributed memory platforms. Our solver symPACK relies on efficient and flexible communication primitives provided by the UPC++ library. Performance evaluation shows good scalability and that symPACK outperforms state-of-the-art parallel distributed memory factorization packages, validating our approach on practical cases.

Jacquelin, Mathias↗

Short ensembles: An Efficient Method for Discerning Climate-relevant Sensitivities in Atmospheric General Circulation Models

This paper explores the feasibility of an experimentation strategy for investigating sensitivities in fast components of atmospheric general circulation models. The basic idea is to replace the traditional serial-in-time long-term climate integrations by representative ensembles of shorter simulations. The key advantage of the proposed method lies in its efficiency: since fewer days of simulation are needed, the computational cost is less, and because individual realizations are independent and can be integrated simultaneously, the new dimension of parallelism can dramatically reduce the turnaround time in benchmark tests, sensitivities studies, and model tuning exercises. The strategy is not appropriate for exploring sensitivity of all model features, but it is very effective in many situations. Two examples are presented using the Community Atmosphere Model version 5. The first example demonstrates that the method is capable of characterizing the model cloud and precipitation sensitivity to time step length. A nudging technique is also applied to an additional set of simulations to help understand the contribution of physics-dynamics interaction to the detected time step sensitivity. In the second example, multiple empirical parameters related to cloud microphysics and aerosol lifecycle are perturbed simultaneously in order to explore which parameters have the largest impact on the simulated global mean top-of-atmosphere radiation balance. Results show that in both examples, short ensembles are able to correctly reproduce the main signals of model sensitivities revealed by traditional long-term climate simulations for fast processes in the climate system. The efficiency of the ensemble method makes it particularly useful for the development of high-resolution, costly and complex climate models.

atmospheric general circulation models↗

Accelerating Simulation for High-Fidelity PV Inverter System Reliability Assessment with High-Performance Computing

The overall cost of photovoltaic (PV) systems has shown a downward trend during the last decade; however, PV inverter failures account for the highest cost of operation and maintenance. To address this, reliability tools with powerful computation and better accuracy are required for the lifetime prediction and degradation evaluation of PV inverters. This paper proposes an event-driven parallel computing-based simulator. The proposed simulator applies high-performance computing techniques and other accessory optimization techniques-including cluster merging, adaptive model updates, and steady-state identification-to make reliability assessments for PV inverters under given input mission profiles and operating conditions with high efficiency and high fidelity. The main idea of the simulator and its workflow are introduced. Then, a demo PV inverter system simulator is implemented, and the speedup of the total simulations of the switching model reaches 123.03 times.

high-performance computing↗

Particle Acceleration during Magnetic Reconnection in a Low-beta Plasma

Magnetic reconnection is a primary mechanism for particle energization in space and astrophysical plasmas. By carrying out two-dimensional (2D) fully kinetic simulations, we study particle acceleration during magnetic reconnection in plasmas with different plasma β (the ratio between the thermal pressure and the magnetic pressure). For the high-β cases, we do not observe significant particle acceleration. In the low-β regime (β<0.1), we find that reconnection is efficient at energizing both electrons and ions. While the distribution of accelerated particles integrated over the whole simulation box appears highly non-thermal, it is actually the superposition of a series of distributions in different sectors of a 2D magnetic island. Each of those distributions has only a small non-thermal component compared with its thermal core. By tracking a large number of particles, we show that particles get energized in X-line regions, contracting magnetic islands, and magnetic island coalescence regions. We obtain the particle energization rate j⋅E by averaging over particle drift motions and find that it agrees well with the particle kinetic energy change. We quantify the contribution of curvature drift, gradient drift, polarization drift, magnetization, non-gyrotropic effect, and parallel electric field in different acceleration regions. We find that the major energization is due to particle curvature drift along the motional electric field. The other particle motions contribute less but may become important in different acceleration regions. The highly efficient particle energization in low-β plasmas may help us understand the strong particle energization in solar flares and accretion disk coronae.

79 ASTRONOMY AND ASTROPHYSICS↗

Regional heritability advanced complex trait analysis for GPU and traditional parallel architectures

Motivation: Quantification of the contribution of genetic variation to phenotypic variation for complex traits becomes increasingly computationally demanding with increasing numbers of single-nucleotide polymorphisms and individuals. To meet the challenges in making feasible large-scale studies, we present the REgional heritability advanced complex trait analysis software. Adapted from advanced complex trait analysis (and, in turn, genome-wide complex trait analysis), it is tailored to exploit the parallelism present in modern traditional and graphics processing unit (GPU)-accelerated machines, from workstations to supercomputers. Results: We adapt the genetic relationship matrix estimation algorithm to remove limitations on memory, allowing the analysis of large datasets. Further, we build on this to develop a version of the code able to efficiently exploit GPU-accelerated systems for both the genetic relationship matrix and REstricted maximum likelihood (REML) parts of the analysis, offering substantial speedup over the traditional central processing unit version. We develop the ability to analyze multiple small regions of the genome across multiple compute nodes in parallel, following the ‘regional heritability’ approach. We demonstrate the new software using 1024 GPUs in parallel on one of the world’s fastest supercomputers. Availability: The code is freely available at http://www.epcc.ed.ac.uk/software-products

59 BASIC BIOLOGICAL SCIENCES↗

Running Parallel Discrete Event Simulators on Sierra

In this proposal we consider porting the ROSS/Charm++ simulator and the discrete event models that run under its control so that they run on the Sierra architecture and make efficient use of the Volta GPUs.

97 MATHEMATICS AND COMPUTING↗

Efficient scalable algorithms for hierarchically semiseparable matrices

Hierarchically semiseparable (HSS) matrix algorithms are emerging techniques in constructing the superfast direct solvers for both dense and sparse linear systems. Here, we develope a set of novel parallel algorithms for the key HSS operations that are used for solving large linear systems. These include the parallel rank-revealing QR factorization, the HSS constructions with hierarchical compression, the ULV HSS factorization, and the HSS solutions. The HSS tree based parallelism is fully exploited at the coarse level. The BLACS and ScaLAPACK libraries are used to facilitate the parallel dense kernel operations at the ne-grained level. We have appplied our new parallel HSS-embedded multifrontal solver to the anisotropic Helmholtz equations for seismic imaging, and were able to solve a linear system with 6.4 billion unknowns using 4096 processors, in about 20 minutes. The classical multifrontal solver simply failed due to high demand of memory. To our knowledge, this is the first successful demonstration of employing the HSS algorithms in solving the truly large-scale real-world problems. Our parallel strategies can be easily adapted to the parallelization of the other rank structured methods.

01 COAL, LIGNITE, AND PEAT↗

Note: A disposable x-ray camera based on mass produced complementary metal-oxide-semiconductor sensors and single-board computers

We have integrated mass-produced commercial complementary metal-oxide-semiconductor (CMOS) image sensors and off-the-shelf single-board computers into an x-ray camera platform optimized for acquisition of x-ray spectra and radiographs at energies of 2–6 keV. The CMOS sensor and single-board computer are complemented by custom mounting and interface hardware that can be easily acquired from rapid prototyping services. For single-pixel detection events, i.e., events where the deposited energy from one photon is substantially localized in a single pixel, we establish ∼20% quantum efficiency at 2.6 keV with ∼190 eV resolution and a 100 kHz maximum detection rate. The detector platform’s useful intrinsic energy resolution, 5-μm pixel size, ease of use, and obvious potential for parallelization make it a promising candidate for many applications at synchrotron facilities, in laser-heating plasma physics studies, and in laboratory-based x-ray spectrometry.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Universal Design Principles for Cascade Heterojunction Solar Cells with High Fill Factors and Internal Quantum Efficiencies Approaching 100%

Cascade heterojunction (CHJ) organic solar cells have recently emerged as an alternative to conventional bulk heterojunctions and series-connected tandems due to their significant promise for high internal quantum efficiency (IQE) and broad spectral coverage. However, CHJ devices thus far have also exhibited poor fill factor (FF), resulting in minimal enhancements (or even decreases) in power conversion efficiency (PCE) when compared with single heterojunction (SHJ) cells. In this study, the major variables controlling the CHJ maximum power point and FF are determined using a combinatorial approach. By matching the maximum power point voltage (VMPP) of the constituent parallel-connected heterojunctions (subjunctions) and minimizing the injection barriers intrinsic to CHJs, high FF and PCE can be achieved. Optimized CHJ devices are demonstrated with >99% IQE in the interlayer and a 46% increase in PCE compared to a SHJ reference (4.1% versus 2.8%). Devices with a transparent exciton dissociation layer (EDL)/interlayer/acceptor structure are employed, such that each CHJ has absorption efficiency identical to its interlayer/acceptor SHJ counterpart. Using these results, a clear map of performance as a function of material parameters is developed, providing straightforward, universal design rules to guide future engineering of molecules and layer architectures for CHJ organic photovoltaic devices.

36 MATERIALS SCIENCE↗