Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel applications”

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 721 records · Page 40

Scalable Adaptive Multilevel Solvers for Multiphysics Problems

In this project, we carried out many studies on adaptive and parallel multilevel methods for numerical modeling for various applications, including Magnetohydrodynamics (MHD) and complex fluids. We have made significant efforts and advances in adaptive multilevel methods of the multiphysics problems: multigrid methods, adaptive finite element methods, and applications.

97 MATHEMATICS AND COMPUTING↗

Comprehensive Measurement and Analysis of the User-Perceived I/O Performance in a Production Leadership-Class Storage System

With the increase of the scale and intensity of the parallel I/O workloads generated by those scientific applications running on high performance computing facilities, understanding the I/O dynamics, especially the root cause of the I/O performance variability and degradation in HPC environment, have become extremely critical to the HPC community. In this paper, we run extensive I/O measuring tests on a production leadership-class storage system to capture the performance variabilities of large-scale parallel I/O. Analyzing these results and its statistic correlation revealed some valuable insights into the characteristics of the storage system and the root cause of I/O performance variability. Further, we leverage these findings and propose an I/O middleware design refactoring which can improve the performance of the parallel I/O by optimizing the data striping and placement. Our preliminary evaluation results demonstrate the proposed approach can reduce the average per-process write latency by at least 80% and the maximum per-process write latency by at least 20%.

Wan, Lipeng↗

Exploring the use of novel programming models in land surface models

A wide range of programming models are currently under rapid development to meet the needs of application developers looking to work on more complex machines. These models fill a variety of roles. Some look to abstract supercomputer architecture, including both processors and memory, to present a strategy for portable performance across a wide range of machines. Others look to expose concurrency by explicitly constructing task-driven dependency graphs that allow a scheduler to find parallelism. Here we explore the implications for application codes of adopting two such programming models, Kokkos and Legion, one from each class of models. We specifically focus on the software design implications on refactoring existing applications, rather than the performance and performance tuning of these models. We identify a strategy for refactoring the Energy Exascale Earth System Model's Land Surface Model, an extremely complex code for climate applications, and prototype a series of mini-apps that explore the adoption of Kokkos and Legion. In doing this, we identify commonalities across the models, leading to a series of conclusions about application software design and refactoring for the adoption of novel programming models.

Coon, Ethan↗

Techniques for Enabling Highly Efficient Message Passing on Many-Core Architectures

Many-core architecture provides a massively parallel environment with dozens of cores and hundreds of hardware threads. Scientific application programmers are increasingly looking at ways to utilize such large numbers of lightweight cores for various programming models. Efficiently executing these models on massively parallel many-core environments is not easy, however and performance may be degraded in various ways. The first author's doctoral research focuses on exploiting the capabilities of many-core architectures on widely used MPI implementations. While application programmers have studied several approaches to achieve better parallelism and resource sharing, many of those approaches still face communication problems that degrade performance. In the thesis, we investigate the characteristics of MPI on such massively threaded architectures and propose two efficient strategies -- a multi-threaded MPI approach and a process-based asynchronous model -- to optimize MPI communication for modern scientific applications.

Si, Min↗

Scaling soft matter physics to thousands of graphics processing units in parallel

We describe a multi-graphics processing unit (GPU) implementation of the Ludwig application, which specialises in simulating a variety of complex fluids via lattice Boltzmann fluid dynamics coupled to additional physics describing complex fluid constituents. We describe our methodology in augmenting the original central processing unit (CPU) version with GPU functionality in a maintainable fashion. We present several optimisations that maximise performance on the GPU architecture through tuning for the GPU memory hierarchy. We describe how we implement particles within the fluid in such a way to avoid a major diversion of the CPU and GPU codebases, whilst minimising data transfer at each time step. We detail our halo-exchange communication phase for the code, which exploits overlapping to allow efficient parallel scaling to many GPUs. We present results showing that the application demonstrates excellent scaling to at least 8192 GPUs in parallel, the largest system tested at the time of writing. The GPU version (on NVIDIA K20X GPUs) is around 3.5–5 times faster that the CPU version (on fully utilised AMD Opteron 6274 16-core CPUs), comparing equal numbers of CPUs and GPUs.

Computer Science↗

Statistical inference of empirical constituents in partitioned analysis from integral-effect experiments: An application in thermo-mechanical coupling

Rapid advancements in parallel computing over the last two decades have enabled simulations of complex, coupled systems through partitioning. In partitioned analysis, independently developed constituent models communicate, representing dependencies between multiple physical phenomena that occur in the full system. Figure 1 schematically demonstrates a coupled system with two constituent models, each resolving different physical behavior. In this figure, the constituent model, denoted as the “consumer,” relies upon some input parameter that is being provided by the constituent model acting as a “feeder”. The role of the feeder model is to map operating conditions (i.e. those that are stimulating the process) to consumer inputs, thus providing functional inputs to the consumer model*. Problems arise if the feeder model cannot be built–a challenge that is prevalent for highly complex systems in extreme operational conditions that push the limits of our understanding of underlying physical behavior. Often, these are also the situations where separate-effect experiments isolating the physical phenomena are not available; meaning that experimentally determining the unknown constituent behavior is not possible (Bauer and Holland, 1995; Unal et al., 2013), and that integral-effect experiments that reflect the behavior of the complete system tend to be the only available observations. In this paper, the authors advocate for the usefulness of integral-effect experiments in furthering a model developer’s knowledge of the physics principles governing the system behavior of interest.

97 MATHEMATICS AND COMPUTING↗

Molecular underpinnings of ssDNA specificity by Rep HUH-endonucleases and implications for HUH-tag multiplexing and engineering

Abstract Replication initiator proteins (Reps) from the HUH-endonuclease superfamily process specific single-stranded DNA (ssDNA) sequences to initiate rolling circle/hairpin replication in viruses, such as crop ravaging geminiviruses and human disease causing parvoviruses. In biotechnology contexts, Reps are the basis for HUH-tag bioconjugation and a critical adeno-associated virus genome integration tool. We solved the first co-crystal structures of Reps complexed to ssDNA, revealing a key motif for conferring sequence specificity and for anchoring a bent DNA architecture. In combination, we developed a deep sequencing cleavage assay, termed HUH-seq, to interrogate subtleties in Rep specificity and demonstrate how differences can be exploited for multiplexed HUH-tagging. Together, our insights allowed engineering of only four amino acids in a Rep chimera to predictably alter sequence specificity. These results have important implications for modulating viral infections, developing Rep-based genomic integration tools, and enabling massively parallel HUH-tag barcoding and bioconjugation applications.

59 BASIC BIOLOGICAL SCIENCES↗

Optimization of Condensed Matter Physics Application with OpenMP Tasking Model

The Density Matrix Renormalization Group (DMRG++) is a condensed matter physics application used to study superconductivity properties of materials. It’s main computations consist of calculating hamiltonian matrix which requires sparse matrix-vector multiplications. This paper presents task-based parallelization and optimization strategies of the Hamiltonian algorithm. The algorithm is implemented as a mini-application in C++ and parallelized with OpenMP. The optimization leverages tasking features, such as dependencies or priorities included in the OpenMP standard 4.5. The code refactoring targets performance as much as programmability. The optimized version achieves a speedup of 8.0× with 8 threads and 20.5× with 40 threads on a Power9 computing node while reducing the memory consumption to 90 MB with respect to the original code, by adding less than ten OpenMP directives.

Criado, Joel↗

MADmap: A Massively Parallel Maximum-Likelihood Cosmic Microwave Background Map-Maker

MADmap is a software application used to produce maximum-likelihood images of the sky from time-ordered data which include correlated noise, such as those gathered by Cosmic Microwave Background (CMB) experiments. It works efficiently on platforms ranging from small workstations to the most massively parallel supercomputers. Map-making is a critical step in the analysis of all CMB data sets, and the maximum-likelihood approach is the most accurate and widely applicable algorithm; however, it is a computationally challenging task. This challenge will only increase with the next generation of ground-based, balloon-borne and satellite CMB polarization experiments. The faintness of the B-mode signal that these experiments seek to measure requires them to gather enormous data sets. MADmap is already being run on up to O(1011) time samples, O(108) pixels and O(104) cores, with ongoing work to scale to the next generation of data sets and supercomputers. We describe MADmap's algorithm based around a preconditioned conjugate gradient solver, fast Fourier transforms and sparse matrix operations. We highlight MADmap's ability to address problems typically encountered in the analysis of realistic CMB data sets and describe its application to simulations of the Planck and EBEX experiments. The massively parallel and distributed implementation is detailed and scaling complexities are given for the resources required. MADmap is capable of analysing the largest data sets now being collected on computing resources currently available, and we argue that, given Moore's Law, MADmap will be capable of reducing the most massive projected data sets.

79 ASTRONOMY AND ASTROPHYSICS↗

Understanding the use of message passing interface in exascale proxy applications

The Exascale Computing Project (ECP) focuses on the development of future exascale-capable applications. Most ECP applications use the message passing interface (MPI) as their parallel programming model with mini-apps serving as proxies. This paper explores the explicit usage of MPI in such ECP proxy applications. We empirically analyze 14 proxy applications from the ECP Proxy Apps Suite. We use the MPI profiling interface (PMPI) to collect MPI usage patterns in ECP proxy apps. Our analysis shows that a small subset of features from MPI is commonly used in the proxies of exascale-capable applications, even when they reference third-party libraries. Overall, this study is intended to provide a better understanding of the use of MPI in current exascale applications. The findings can help focus software investments made for exascale systems in the MPI middleware including optimization, fault-tolerance, tuning, and hardware-offload.

97 MATHEMATICS AND COMPUTING↗

Transparent redundant computing with MPI.

Extreme-scale parallel systems will require alternative methods for applications to maintain current levels of uninterrupted execution. Redundant computation is one approach to consider, if the benefits of increased resiliency outweigh the cost of consuming additional resources. We describe a transparent redundancy approach for MPI applications and detail two different implementations that provide the ability to tolerate a range of failure scenarios, including loss of application processes and connectivity.We compare these two approaches and show performance results from micro-benchmarks that bound worst-case message passing performance degradation.We propose several enhancements that could lower the overhead of providing resiliency through redundancy.

99 GENERAL AND MISCELLANEOUS↗

Quasielastic neutron scattering in biology: Theory and applications

Neutrons scatter quasielastically from stochastic, diffusive processes, such as overdamped vibrations, localized diffusion and transitions between energy minima. In biological systems, such as proteins and membranes, these relaxation processes are of considerable physical interest. We review here recent methodological advances and applications of quasielastic neutron scattering (QENS) in biology, concentrating on the role of molecular dynamics simulation in generating data with which neutron profiles can be unambiguously interpreted. We examine the use of massively-parallel computers in calculating scattering functions, and the application of Markov state modeling. The decomposition of MD-derived neutron dynamic susceptibilities is described, and the use of this in combination with NMR spectroscopy. We discuss dynamics at very long times, including approximations to the infinite time mean-square displacement and nonequilibrium aspects of single-protein dynamics. Lastly, we examine how neutron scattering and MD can be combined to provide information on lipid nanodomains.

59 BASIC BIOLOGICAL SCIENCES↗

Engineering growth factors for regenerative medicine applications.

Growth factors are important morphogenetic proteins that instruct cell behavior and guide tissue repair and renewal. Although their therapeutic potential holds great promise in regenerative medicine applications, translation of growth factors into clinical treatments has been hindered by limitations including poor protein stability, low recombinant expression yield, and suboptimal efficacy. This review highlights current tools, technologies, and approaches to design integrated and effective growth factor-based therapies for regenerative medicine applications. The first section describes rational and combinatorial protein engineering approaches that have been utilized to improve growth factor stability, expression yield, biodistribution, and serum half-life, or alter their cell trafficking behavior or receptor binding affinity. The second section highlights elegant biomaterial-based systems, inspired by the natural extracellular matrix milieu, that have been developed for effective spatial and temporal delivery of growth factors to cell surface receptors. Although appearing distinct, these two approaches are highly complementary and involve principles of molecular design and engineering to be considered in parallel when developing optimal materials for clinical applications.

Biomaterials↗

Vidyut3d: A Non-Equilibrium Plasma Modeling Tool [SWR-24-101]

Vidyut3d is a massively-parallel plasma-fluid solver for low-temperature plasmas (LTPs) that supports both local field (LFA) and local mean energy (LMEA) approximations, as well as complex gas and surface-phase chemistry. The solver supports 2D and 3D domains, and uses AMReX's adaptive mesh refinement capabilities to increase the grid resolution around complex structures (e.g. streamer heads and sheaths) while maintaining a tractable problem size. Vidyut specializes in simulating various types of gas-phase discharges, as well as plasma/surface interactions and surface chemistry (e.g. for plasma-mediated catalysis applications). The solver also supports hybrid CPU/GPU parallelization strategies, and has demonstrated excellent scaling on various HPC architectures for problem sizes consisting of O(100 M) control volumes.

Sitaraman, Hariswaran↗

Exploring MPI Communication Models for Graph Applications Using Graph Matching as a Case Study

Traditional implementations of parallel graph operations on distributed memory platforms are written using Message Passing Interface (MPI) point to point communication primitives such as Send-Recv (blocking and nonblocking). Apart from this classical model, the MPI community has over the years added other communication models; however, their suitability for handling the irregular traffic workloads typical of graph operations remain comparatively less explored. Our aim in this paper is to study these relatively underutilized communication models of MPI for graph applications. More specifically, we evaluate MPI’s one-sided programming, or Remote Memory Access (RMA), and nearest neighborhood collectives using a process graph topology. There are features in these newer models that are intended to better map to irregular communication patterns, as exemplified in graph algorithms. As a concrete application for our case study, we use distributed memory implementations of a weighted graph matching algorithm to investigate performances of MPI-3 RMA and neighborhood collective operations compared to nonblocking Send- Recv. A matching in a graph is a subset of edges such that no two matched edges are incident on the same vertex. A maximum weight matching is a matching of maximum weight computed as the sum of the weights of matched edges. Algorithms for optimal matching have quadratic complexity and are difficult to parallelize. More importantly, execution of graph matching is dominated by high volume of irregular memory accesses, making it an ideal candidate for studying the effects of various MPI communication models on graph applications at scale. Our neighborhood collectives and RMA implementations yields up to 6× speedup over traditional optimized nonblocking Send-Recv implementations, on thousands of cores of the NERSC Cori supercomputer. We believe the lessons learned from this case study can be adopted to benefit a wider range of graph application codes.

Ghosh, Sayan↗

Application Note: Power Grid Modeling With Xyce™

This application note describes how to model steady-state power flows and transient events in electric power grids with the SPICE-compatible Xyce™ Parallel Electronic Simulator developed at Sandia National Labs. This application notes provides a brief tutorial on the basic devices (branches, bus shunts, transformers and generators) found in power grids. The focus is on the features supported and assumptions made by the Xyce models for power grid elements. It then provides a detailed explanation, including working Xyce netlists, for simulating some simple power grid examples such as the IEEE 14-bus test case.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Exploiting variability for energy optimization of parallel programs

Here in this paper we present optimizations that use DVFS mechanisms to reduce the total energy usage in scientific applications. Our main insight is that noise is intrinsic to large scale parallel executions and it appears whenever shared resources are contended. The presence of noise allows us to identify and manipulate any program regions amenable to DVFS. When compared to previous energy optimizations that make per core decisions using predictions of the running time, our scheme uses a qualitative approach to recognize the signature of executions amenable to DVFS. By recognizing the "shape of variability" we can optimize codes with highly dynamic behavior, which pose challenges to all existing DVFS techniques. We validate our approach using offline and online analyses for one-sided and two-sided communication paradigms. We have applied our methods to NWChem, and we show best case improvements in energy use of 12% at no loss in performance when using online optimizations running on 720 Haswell cores with one-sided communication. With NWChem on MPI two-sided and offline analysis, capturing the initialization, we find energy savings of up to 20%, with less than 1% performance cost.

97 MATHEMATICS AND COMPUTING↗

A low-communication, parallel algorithm for solving PDEs based on range decomposition

This work proposes a new, low-communication algorithm for solving PDEs on massively parallel computers. The range decomposition (RD) algorithm exposes coarse-grain parallelism by applying nested iteration and adaptive mesh refinement locally before performing a global communication step. Just a few such steps are observed to be sufficient to obtain accuracy within a small multiple of discretization error. The target applications are petascale and exascale machines, where hierarchical parallelism is required and traditional parallel numerical PDE communication patterns are costly because of message latency. The RD algorithm uses a partition of unity to equally distribute the error, and thus, the work. The computational advantages of this approach are that the decomposed problems can be solved in parallel without any communication until the partitioned solutions are summed. This offers potential advantages in the paradigm of expensive communication but very cheap computation. This paper introduces the method and explains the details of the communication step. Two performance models are developed, showing that the latency cost associated with a traditional parallel implementation of nested iteration is proportional to log(P) 2 , whereas the RD method reduces the communication latency to log(P), while maintaining similar bandwidth costs. Numerical results for two problems, Laplace and advection diffusion, demonstrate the enhanced performance, and a heuristic argument explains why the method converges quickly.

97 MATHEMATICS AND COMPUTING↗