Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel processors”

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 1,729 records · Page 96

DataStager: scalable data staging services for petascale applications

Known challenges for petascale machines are that (1) the costs of I/O for high performance applications can be substantial, especially for output tasks like checkpointing, and (2) noise from I/O actions can inject undesirable delays into the runtimes of such codes on individual compute nodes. This paper introduces the flexible 'DataStager' framework for data staging and alternative services within that jointly address (1) and (2). Data staging services moving output data from compute nodes to staging or I/O nodes prior to storage are used to reduce I/O overheads on applications total processing times, and explicit management of data staging offers reduced perturbation when extracting output data from a petascale machine's compute partition. Experimental evaluations of DataStager on the Cray XT machine at Oak Ridge National Laboratory establish both the necessity of intelligent data staging and the high performance of our approach, using the GTC fusion modeling code and benchmarks running on 1000+ processors.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

The Impact of IBM Cell Technology on the Programming Paradigm in the Context of Computer Systems for Climate and Weather Models

The call for ever-increasing model resolutions and physical processes in climate and weather models demands a continual increase in computing power. The IBM Cell processor's order-of-magnitude peak performance increase over conventional processors makes it very attractive to fulfill this requirement. However, the Cell's characteristics, 256KB local memory per SPE and the new low-level communication mechanism, make it very challenging to port an application. As a trial, we selected the solar radiation component of the NASA GEOS-5 climate model, which: (1) is representative of column physics components (half the total computational time), (2) has an extremely high computational intensity: the ratio of computational load to main memory transfers, and (3) exhibits embarrassingly parallel column computations. In this paper, we converted the baseline code (single-precision Fortran) to C and ported it to an IBM BladeCenter QS20. For performance, we manually SIMDize four independent columns and include several unrolling optimizations. Our results show that when compared with the baseline implementation running on one core of Intel's Xeon Woodcrest, Dempsey, and Itanium2, the Cell is approximately 8.8x, 11.6x, and 12.8x faster, respectively. Our preliminary analysis shows that the Cell can also accelerate the dynamics component (~;;25percent total computational time). We believe these dramatic performance improvements make the Cell processor very competitive as an accelerator.

97 MATHEMATICS AND COMPUTING↗

The use of imprecise processing to improve accuracy in weather and climate prediction

The use of stochastic processing hardware and low precision arithmetic in atmospheric models is investigated. Stochastic processors allow hardware-induced faults in calculations, sacrificing bit-reproducibility and precision in exchange for improvements in performance and potentially accuracy of forecasts, due to a reduction in power consumption that could allow higher resolution. A similar trade-off is achieved using low precision arithmetic, with improvements in computation and communication speed and savings in storage and memory requirements. As high-performance computing becomes more massively parallel and power intensive, these two approaches may be important stepping stones in the pursuit of global cloud-resolving atmospheric modelling. The impact of both hardware induced faults and low precision arithmetic is tested using the Lorenz '96 model and the dynamical core of a global atmosphere model. In the Lorenz '96 model there is a natural scale separation; the spectral discretisation used in the dynamical core also allows large and small scale dynamics to be treated separately within the code. Such scale separation allows the impact of lower-accuracy arithmetic to be restricted to components close to the truncation scales and hence close to the necessarily inexact parametrised representations of unresolved processes. By contrast, the larger scales are calculated using high precision deterministic arithmetic. Hardware faults from stochastic processors are emulated using a bit-flip model with different fault rates. Our simulations show that both approaches to inexact calculations do not substantially affect the large scale behaviour, provided they are restricted to act only on smaller scales. By contrast, results from the Lorenz '96 simulations are superior when small scales are calculated on an emulated stochastic processor than when those small scales are parametrised. This suggests that inexact calculations at the small scale could reduce computation and power costs without adversely affecting the quality of the simulations. This would allow higher resolution models to be run at the same computational cost.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Scalable Modeling of Carbon Tetrachloride Migration at the Hanford Site Using the STOMP Simulator

Numerical simulation has been applied in support of the U.S. Department of Energy’s (DOE’s) efforts to characterize the nature and distribution of carbon tetrachloride in the deep vadose zone at the Hanford site, near Richland, Washington. Three-dimensional computational domains were used, with layered and heterogeneous distributions of soil properties, in this numerical investigation into the vertical and lateral distribution of carbon tetrachloride beneath it release point (216-Z-9 trench) and the effects of soil vapor extraction process. The complexity of the modeled physical processes, namely, the nonlinearities associated with multifluid subsurface flow, including phase transitions and hysteresis in the relative permeability-saturation-capillary pressure functions, limits the grid resolution when executed using single processor computers. To achieve higher grid resolutions and acceptable detail in the subsurface distribution and remediation of carbon tetrachloride, execution on multiple processors was required. This paper describes and demonstrates a scalable implementation of a multifluid subsurface flow and transport with capabilities for volatile organic compounds, residual nonaqueous phase liquid formation in the vadose zone, and soil vapor extraction, using multiple wells. Developing scientific software for execution on parallel computers has unique challenges. The guiding objectives for developing this scalable code were to keep the source coding readable and modifiable by subsurface scientists, allow for both sequential and scalable processing, depend on domain scientists for code parallelization and scalable linear system solvers.

216-Z-9 Trench↗

SHARP Assembly-Scale Multiphysics Demonstration Simulations

The NEAMS Reactor Product Line effort aims to develop an integrated multiphysics simulation capability for the design and analysis of future generations of nuclear power plants. The Reactor Product Line code suite’s multi-resolution hierarchy is being designed to ultimately span the full range of length and time scales present in relevant reactor design and safety analyses, as well as scale from desktop to petaflop computing platforms. In this report, we describe our efforts to integrate thermal/hydraulics, neutronics, and structural mechanics modeling codes to perform coupled analysis of a representative fast sodium-cooled reactor fuel assembly. Over the past five years, the Reactor Product Line effort has developed high-fidelity single-physics codes for neutron transport modeling, in the PROTEUS code, and Large Eddy Simulation-based thermal/fluid modeling in the Nek5000 code. Both these codes have been exercised on over 100,000 processors of the IBM Blue Gene/P. The Diablo code has been used to perform structural mechanics and thermo-mechanical modeling. MOAB, the Reactor Geometry Generator (RGG), and MeshKit have been developed to generate and manipulate mesh and mesh-based data, in both serial and parallel environments. These tools together form a strong basis on which to build a multi-physics modeling capability. The goal of developing such a tool is to perform multi-physics neutronics, thermal/fluid, and structural mechanics modeling of the components inside a reactor core. While the focus of this report is on modeling a fast sodium-cooled reactor, it is also the goal that this simulation tool be useful for most reactor types. Here we report on the integration of PROTEUS and Nek5000 into the NEAMS framework. Both codes have been modified to read and write data from and to the MOAB library, in both serial and parallel environments. We focused on broad support for as many original physics code options as possible, rather than focusing only on near-term needs. Both codes were demonstrated through this interface on a relatively simple, seven-pin hexagonal fuel assembly. To support coupled analysis, the new multi-physics driver library COUPE was developed; this library provides flexible support for both loose and tight coupling, and can interface flexibly with codes in multiple programming languages. To couple results between the separate meshes used in each physics, the driver, PROTEUS, and Nek5000 all interface with the MBCoupler solution transfer tool provided with MOAB. Together, these tools form the basis of a new, coupled neutronics and thermal/fluid modeling code, that works in both serial and parallel. In a related effort, a driver in the Diablo code was used to access Nek5000-computed results written to a MOAB database, as well as the temperature data used to drive structural mechanics computations.

Tautges, Timothy J.↗

Advances in Parallelization for Large Scale Oct-Tree Mesh Generation

Despite great advancements in the parallelization of numerical simulation codes over the last 20 years, it is still common to perform grid generation in serial. Generating large scale grids in serial often requires using special "grid generation" compute machines that can have more than ten times the memory of average machines. While some parallel mesh generation techniques have been proposed, generating very large meshes for LES or aeroacoustic simulations is still a challenging problem. An automated method for the parallel generation of very large scale off-body hierarchical meshes is presented here. This work enables large scale parallel generation of off-body meshes by using a novel combination of parallel grid generation techniques and a hybrid "top down" and "bottom up" oct-tree method. Meshes are generated using hardware commonly found in parallel compute clusters. The capability to generate very large meshes is demonstrated by the generation of off-body meshes surrounding complex aerospace geometries. Results are shown including a one billion cell mesh generated around a Predator Unmanned Aerial Vehicle geometry, which was generated on 64 processors in under 45 minutes.

O'Connell, Matthew↗

Parallel Processing of Broad-Band PPM Signals

A parallel-processing algorithm and a hardware architecture to implement the algorithm have been devised for timeslot synchronization in the reception of pulse-position-modulated (PPM) optical or radio signals. As in the cases of some prior algorithms and architectures for parallel, discrete-time, digital processing of signals other than PPM, an incoming broadband signal is divided into multiple parallel narrower-band signals by means of sub-sampling and filtering. The number of parallel streams is chosen so that the frequency content of the narrower-band signals is low enough to enable processing by relatively-low speed complementary metal oxide semiconductor (CMOS) electronic circuitry. The algorithm and architecture are intended to satisfy requirements for time-varying time-slot synchronization and post-detection filtering, with correction of timing errors independent of estimation of timing errors. They are also intended to afford flexibility for dynamic reconfiguration and upgrading. The architecture is implemented in a reconfigurable CMOS processor in the form of a field-programmable gate array. The algorithm and its hardware implementation incorporate three separate time-varying filter banks for three distinct functions: correction of sub-sample timing errors, post-detection filtering, and post-detection estimation of timing errors. The design of the filter bank for correction of timing errors, the method of estimating timing errors, and the design of a feedback-loop filter are governed by a host of parameters, the most critical one, with regard to processing very broadband signals with CMOS hardware, being the number of parallel streams (equivalently, the rate-reduction parameter).

Gray, Andrew↗

Parallel algorithms for geometric connected component labeling on a hypercube multiprocessor

Different algorithms for the geometric connected component labeling (GCCL) problem are defined each of which involves d stages of message passing, for a d-dimensional hypercube. The major idea is that in each stage a hypercube multiprocessor increases its knowledge of domain. The algorithms under consideration include the QUAD algorithm for small number of processors and the Overlap Quad algorithm for large number of processors, subject to the locality of the connected sets. These algorithms differ in their run time, memory requirements, and message complexity. They were implemented on an Intel iPSC2/D4/MX hypercube.

Belkhale, K. P.↗

FOX: A Fault Oblivious Extreme-Scale Execution Environment

The FOX project, funded under the ASCR X-stack I program, developed systems software and runtime libraries for a new approach to the data and work distribution for massively parallel, fault oblivious application execution. Our work was motivated by the premise that exascale computing systems will provide a thousand-fold increase in parallelism and a proportional increase in failure rate relative to today’s machines. To deliver the capability of exascale hardware, the systems software must provide the infrastructure to support existing applications while simultaneously enabling efficient execution of new programming models that naturally express dynamic, adaptive, irregular computation; coupled simulations; and massive data analysis in a highly unreliable hardware environment with billions of threads of execution. Our OS research has prototyped new methods to provide efficient resource sharing, synchronization, and protection in a many-core compute node. We have experimented with alternative task/dataflow programming models and shown scalability in some cases to hundreds of thousands of cores. Much of our software is in active development through open source projects. Concepts from FOX are being pursued in next generation exascale operating systems. Our OS work focused on adaptive, application tailored OS services optimized for multi → many core processors. We developed a new operating system NIX that supports role-based allocation of cores to processes which was released to open source. We contributed to the IBM FusedOS project, which promoted the concept of latency-optimized and throughput-optimized cores. We built a task queue library based on distributed, fault tolerant key-value store and identified scaling issues. A second fault tolerant task parallel library was developed, based on the Linda tuple space model, that used low level interconnect primitives for optimized communication. We designed fault tolerance mechanisms for task parallel computations employing work stealing for load balancing that scaled to the largest existing supercomputers. Finally, we implemented the Elastic Building Blocks runtime, a library to manage object-oriented distributed software components. To support the research, we won two INCITE awards for time on Intrepid (BG/P) and Mira (BG/Q). Much of our work has had impact in the OS and runtime community through the ASCR Exascale OS/R workshop and report, leading to the research agenda of the Exascale OS/R program. Our project was, however, also affected by attrition of multiple PIs. While the PIs continued to participate and offer guidance as time permitted, losing these key individuals was unfortunate both for the project and for the DOE HPC community.

97 MATHEMATICS AND COMPUTING↗

Neutronics Calculation of Fast Reactor Using Modern Computing Systems

The possibility of using the LUCKY-A three-dimensional multiprocessor transport code as a neutronics module for the COREMELT integrated non-stationary code for performing safety analysis, which is used for fast sodium reactor safety validation, is discussed. In this application, more stringent requirements are imposed on the computing speed of the LUCKY-A code. To test LUCKY-A possibilities, a typical computational problem is examined in a formulation characteristic for the proposed use of the code. The possibilities of the code for parallelized calculations are shown, and a comparison is made with calculations performed with the MMKKENO and TRIGEX codes. The efficacy of a parallel computational process is investigated as a function of the number of computing kernels used.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Gyrokinetic particle-in-cell optimization on emerging multi- and manycore platforms

The next decade of high-performance computing (HPC) systems will see a rapid evolution and divergence of multi- and manycore architectures as power and cooling constraints limit increases in microprocessor clock speeds. Understanding efficient optimization methodologies on diverse multicore designs in the context of demanding numerical methods is one of the greatest challenges faced today by the HPC community. In this paper, we examine the efficient multicore optimization of GTC, a petascale gyrokinetic toroidal fusion code for studying plasma microturbulence in tokamak devices. For GTC’s key computational components (charge deposition and particle push), we explore efficient parallelization strategies across a broad range of emerging multicore designs, including the recently-released Intel Nehalem-EX, the AMD Opteron Istanbul, and the highly multithreaded Sun UltraSparc T2+. We also present the first study on tuning gyrokinetic particle-in-cell (PIC) algorithms for graphics processors, using the NVIDIA C2050 (Fermi). Our work discusses several novel optimization approaches for gyrokinetic PIC, including mixed-precision computation, particle binning and decomposition strategies, grid replication, SIMDized atomic floating-point operations, and effective GPU texture memory utilization. Overall, we achieve significant performance improvements of 1.3–4.7× on these complex PIC kernels, despite the inherent challenges of data dependency and locality. Finally, our work also points to several architectural and programming features that could significantly enhance PIC performance and productivity on next-generation architectures.

97 MATHEMATICS AND COMPUTING↗

Neural network methods for radiation detectors and imaging

Recent advances in image data proccesing through deep learning allow for new optimization and performance-enhancement schemes for radiation detectors and imaging hardware. This enables radiation experiments, which includes photon sciences in synchrotron and X-ray free electron lasers as a subclass, through data-endowed artificial intelligence. We give an overview of data generation at photon sources, deep learning-based methods for image processing tasks, and hardware solutions for deep learning acceleration. Most existing deep learning approaches are trained offline, typically using large amounts of computational resources. However, once trained, DNNs can achieve fast inference speeds and can be deployed to edge devices. A new trend is edge computing with less energy consumption (hundreds of watts or less) and real-time analysis potential. While popularly used for edge computing, electronic-based hardware accelerators ranging from general purpose processors such as central processing units (CPUs) to application-specific integrated circuits (ASICs) are constantly reaching performance limits in latency, energy consumption, and other physical constraints. These limits give rise to next-generation analog neuromorhpic hardware platforms, such as optical neural networks (ONNs), for high parallel, low latency, and low energy computing to boost deep learning acceleration (LA-UR-23-32395).

edge computing↗

Mapping unstructured grid problems to the connection machine

We present a highly parallel graph mapping technique that enables one to solve unstructured grid problems on massively parallel computers. Many implicit and explicit methods for solving discretizated partial differential equations require each point in the discretization to exchange data with its neighboring points every time step or iteration. The time spent communicating can limit the high performance promised by massively parallel computing. To eliminate this bottleneck, we map the graph of the irregular problem to the graph representing the interconnection topology of the computer such that the sum of the distances that the messages travel is minimized. We show that, in comparison to a naive assignment of processors, our heuristic mapping algorithm significantly reduces the communication time on the Connection Machine, CM-2.

Hammond, Steven W.↗

Model-driven mapping onto distributed memory parallel computers

The author addresses the problem of exploiting the parallelism available in a program to efficiently employ the resources of the target machine in the context of building a mapping compiler for a distributed memory parallel machine. He demonstrates the effectiveness of using execution models to select the best mapping technique from among those available for a given program segment on a particular machine. Through analysis of the execution models for several mapping techniques for one class of programs on a linear processor array, it is shown that selecting the best technique for a particular program instance can make a significant difference in performance. On the other hand, the results of benchmarks from a mapping compiler for the Warp systolic array machine show that the execution models considered are accurate enough to select the best mapping technique for a given program.

Sussman, Alan↗

Parallel and Streaming Generation of Ghost Data for Structured Grids

Parallel simulations decompose large domains into many blocks. A fundamental requirement for subsequent parallel analysis and visualization is the presence of ghost data that supplements each block with a layer of adjacent data elements from neighboring blocks. The standard approach for generating ghost data requires all blocks to be in memory at once. This becomes impractical when there are fewer processors - and thus less aggregate memory - available for analysis than for simulation. We describe an algorithm for generating ghost data for structured grids that uses many fewer processors than previously possible. Our algorithm stores as little as one block per processor in memory and can run on as few processors as are available (possibly just one). The key idea is to slightly change the size of the original blocks by declaring parts of them to be ghost data, and by later padding adjacent blocks with this data.

99 GENERAL AND MISCELLANEOUS↗

Solution of structural analysis problems on a parallel computer

The problems of a blade-stiffened panel with a hole subjected to compression, and a deployable space mast subjected to tip loads, are treated through the application of FEM to model generation followed by the solution of a linear system of equations. Direct and iterative approaches to the solution of the linear systems are solved in turn; for the panel problems using varying numbers of processors, the incomplete Cholesky-conjugate gradient method was the fastest iterative method on all but two instances in which the number of processors was large.

Storaasli, Olaf↗

Virtual Cable Impedance based Load Sharing in a Microgrid for Parallel Connected Grid Forming Converters

This paper presents a novel approach to power sharing between direct connected grid-forming converters, utilizing virtual cable impedance and droop-based outer loop control. To enhance stability, resistive droop is implemented, while virtual cable impedance with non-zero resistive and inductive components ensures improved power sharing. The inner loop controller employs a Lyapunov energy function to achieve superior dynamic performance. The proposed control architecture is validated through comprehensive modeling and real-time processor-in-the-loop simulations, demonstrating its robustness and efficiency under various operating conditions. The results highlight the potential of this control strategy to improve the reliability and efficiency of renewable energy systems. Additionally, a comparative analysis with traditional methods underscores the advantages of the proposed approach in terms of stability and performance. The proposed control architecture offers a scalable and flexible solution for grid-forming converters, enabling seamless integration of renewable energy sources. Its robustness and adaptability make it an attractive solution for real-world applications. Furthermore, the approach can be extended to other power electronic systems, enhancing overall system performance and efficiency. By providing a reliable and efficient control strategy, this paper contributes to the advancement of renewable energy systems and their adoption in the energy sector. The proposed control strategy has far-reaching implications for the widespread adoption of renewable energy sources, enabling a more sustainable and efficient energy future. The overall system is modeled in MATLAB/Simulink and PLECS software domain.

grid forming converters (GFM)↗

NWChem: Quantum Chemistry Simulations at Scale

Methods based on quantum mechanics equations have been developed since the 1930's with the purpose of accurately studying the electronic structure of molecules. However, it is only during the last two decades that intense development of new computational algorithms has opened the possibility of performing accurate simulations of challenging molecular processes with high-order many-body methods. A wealth of evidence indicates that the proper inclusion of instantaneous interactions between electrons (or the so-called electron correlation effects) is indispensable for the accurate characterization of chemical reactivity, molecular properties, and interactions of light with matter. The availability of reliable methods for benchmarking of medium-size molecular systems provides also a unique chance to propagate high-level accuracy across spatial scales through the multiscale methodologies. Some of these methods have potential to utilize computational resources in an effi*cient way since they are characterized by high numerical complexity and appropriate level of data granularity, which can be effi*ciently distributed over multi-processor architectures. The broad spectrum of coupled cluster (CC) methods falls into this class of methodologies. Several recent CC implementations clearly demonstrated the scalability of CC formalisms on architectures composed of hundreds thousand computational cores. In this context NWChem provides a collection of Tensor Contraction Engine (TCE) generated parallel implementations of various coupled cluster methods capable of taking advantage of many thousand of cores on leadership class parallel architectures.

chemistry↗