Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallelization”

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 685 records · Page 38

Collective Prefetching for Parallel I/O Systems

Data prefetching can be beneficial for improving parallel I/O system performance, but the amount of benefit depends on how efficiently and swiftly prefetches can be done. In this study, we propose a new prefetching strategy, called collective prefetching. The idea is to exploit the correlation among I/O accesses of multiple processes of a parallel application and carry out prefetches collectively, instead of the traditional strategy of carrying out prefetches by each process individually. The rationale behind this new collective prefetching strategy is that the concurrent processes of the same parallel application have strong correlation with respect to their I/O requests. We present the idea, initial design and implementation of the new collective prefetching strategy in this study. The preliminary experimental results show that this new collective prefetching strategy holds promise for improving parallel I/O performance.

Chen, Yong↗

A Scalable Parallel Wideband MLFMA for Efficient Electromagnetic Simulations on Large Scale Clusters

The development of the multilevel fast multipole algorithm (MLFMA) and its multiscale variants have enabled the use of integral equation (IE) based solvers to compute scattering from complicated structures. Development of scalable parallel algorithms, to extend the reach of these solvers, has been a topic of intense research for about a decade. In this paper, we present a new algorithm for parallel implementation of IE solver that is augmented with a wideband MLFMA and scalable on large number of processors. The wideband MLFMA employed here, to handle multiscale problems, is a hybrid combination of the accelerated Cartesian expansion (ACE) and the classical MLFMA. The salient feature of the presented parallel algorithm is that it is implicitly load balanced and exhibits higher performance. This is achieved by developing a strategy to partition the MLFMA tree, and hence the associated computations, in a self-similar fashion among the parallel processors. As detailed in the paper, the algorithm employs both spatial and direction partitioning approaches in a flexible manner to ensure scalable performance. Plethora of results are presented here to exhibit the scalability of this algorithm on 512 and more processors.

99 GENERAL AND MISCELLANEOUS↗

Reduced-Order Aggregate Model for Parallel-Connected Single-Phase Inverters

This paper outlines a reduced-order aggregate dynamical model for parallel-connected single-phase grid-connected inverters. For each inverter, we place no restrictions on the converter topology and merely assume that the ac-side switch-averaged voltage can be controlled via pulsewidth modulation. The ac output of each inverter interfaces through an LCL filter to the grid. The closed-loop system contains a phase locked loop for grid synchronization, and real- and reactive-power control are realized with inner and outer PI current- and power-control loops. We derive a necessary and sufficient set of parametric relationships to ensure that a reduced-order aggregated state-space model for an arbitrary number of such paralleled inverters has the same model order and structure as any single inverter. We also present reduced-order models for the settings where the real- and reactive-power setpoints are different and where the inverters have different power ratings. We anticipate the proposed model being useful in analyzing the dynamics of large collections of parallel-connected inverters with minimal computational complexity. The aggregate model is validated against measurements obtained from a multi-inverter experimental setup consisting of three 750-VA paralleled grid-connected inverters, hence establishing robustness of the analytical result to parametric variations seen in practice.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Accelerating Bilevel Optimization With Hierarchical Many-Threaded Parallel Differential Evolution

Bilevel optimization is encountered in many relevant real-world applications. The main feature of this type of problem is that an upper-level optimization problem is constrained by a nested lower-level optimization problem. Because of this nested structure, bilevel problems (BLPs) are usually computationally expensive to solve. Differential evolution (DE) has demonstrated promising results in solving BLPs of relatively small scales. As the problem scale increases, the decision space becomes intrinsically larger, requiring a growing number of function evaluations for the method to work properly. In this context, heavy parallelization and high-performance computing techniques are indispensable to enable the resolution of more complex and challenging optimization problems. Hence, we propose a hierarchical many-threaded parallel DE approach for BLPs, where both levels are parallelized. The computational experiments demonstrate that the parallel implementation achieved runtime speeds ranging from 44 to 2559 times faster than the sequential version on a well-known scalable SMD benchmark test problem when executed on an NVIDIA A100 GPU. The findings indicate that the algorithm’s convergence is strongly influenced by the number of both upper- and lower-level generations. Moreover, the success of experiments with large-scale problems is closely linked to the choice of small population sizes.

Dufek, Amanda S↗

pGraph: Efficient Parallel Construction of Large-Scale Protein Sequence Homology Graphs

Detecting sequence homology between protein sequences is a fundamental problem in computational molecular biology, with a pervasive application in nearly all analyses that aim to structurally and functionally characterize protein molecules. While detecting the homology between two protein sequences is relatively inexpensive, detecting pairwise homology for a large number of protein sequences can become computationally prohibitive for modern inputs, often requiring millions of CPU hours. Yet, there is currently no robust support to parallelize this kernel. In this paper, we identify the key characteristics that make this problemparticularly hard to parallelize, and then propose a new parallel algorithm that is suited for detecting homology on large data sets using distributed memory parallel computers. Our method, called pGraph, is a novel hybrid between the hierarchical multiple-master/worker model and producer-consumer model, and is designed to break the irregularities imposed by alignment computation and work generation. Experimental results show that pGraph achieves linear scaling on a 2,048 processor distributed memory cluster for a wide range of inputs ranging from as small as 20,000 sequences to 2,560,000 sequences. In addition to demonstrating strong scaling, we present an extensive report on the performance of the various system components and related parametric studies.

Environmental Molecular Sciences Laboratory↗

Planetary normal mode computation: Parallel algorithms, performance and reproducibility

This report is an extension of work entitled “Computing planetary interior normal modes with a highly parallel polynomial filtering eigensolver.” by Shi et al., [1] originally presented at the SC18 conference. A highly parallel polynomial filtered eigensolver was developed and exploited to calculate the planetary normal modes. The proposed method is ideally suited for computing interior eigenpairs for large-scale eigenvalue problems as it greatly enhances memory and computational efficiency. In this article, the second-order finite element method is used to further improve the accuracy as only the first-order finite element method was deployed in the previous work. The parallel algorithm, its parallel performance up to 20k processors, and the great computational accuracy are illustrated. The reproducibility of the previous work was successfully performed on the Student Cluster Competition at the SC19 conference by several participant teams using a completely different Mars-model dataset on different clusters. Both weak and strong scaling performances of the reproducibility by the participant teams were impressive and encouraging. The analysis and reflection of their results are demonstrated and future direction is discussed.

97 MATHEMATICS AND COMPUTING↗

Heterodyne frequency-domain multispectral diffuse optical tomography of breast cancer in the parallel-plane transmission geometry

Purpose: The authors introduce a state-of-the-art all-optical clinical diffuse optical tomography (DOT) imaging instrument which collects spatially dense, multispectral, frequency-domain breast data in the parallel-plate geometry. Methods: The instrument utilizes a CCD-based heterodyne detection scheme that permits massively parallel detection of diffuse photon density wave amplitude and phase for a large number of source–detector pairs (10{sup 6}). The stand-alone clinical DOT instrument thus offers high spatial resolution with reduced crosstalk between absorption and scattering. Other novel features include a fringe profilometry system for breast boundary segmentation, real-time data normalization, and a patient bed design which permits both axial and sagittal breast measurements. Results: The authors validated the instrument using tissue simulating phantoms with two different chromophore-containing targets and one scattering target. The authors also demonstrated the instrument in a case study breast cancer patient; the reconstructed 3D image of endogenous chromophores and scattering gave tumor localization in agreement with MRI. Conclusions: Imaging with a novel parallel-plate DOT breast imager that employs highly parallel, high-resolution CCD detection in the frequency-domain was demonstrated.

60 APPLIED LIFE SCIENCES↗

TU-AB-BRC-12: Optimized Parallel MonteCarlo Dose Calculations for Secondary MU Checks

Purpose: Secondary MU checks are an important tool used during a physics review of a treatment plan. Commercial software packages offer varying degrees of theoretical dose calculation accuracy, depending on the modality involved. Dose calculations of VMAT plans are especially prone to error due to the large approximations involved. Monte Carlo (MC) methods are not commonly used due to their long run times. We investigated two methods to increase the computational efficiency of MC dose simulations with the BEAMnrc code. Distributed computing resources, along with optimized code compilation, will allow for accurate and efficient VMAT dose calculations. Methods: The BEAMnrc package was installed on a high performance computing cluster accessible to our clinic. MATLAB and PYTHON scripts were developed to convert a clinical VMAT DICOM plan into BEAMnrc input files. The BEAMnrc installation was optimized by running the VMAT simulations through profiling tools which indicated the behavior of the constituent routines in the code, e.g. the bremsstrahlung splitting routine, and the specified random number generator. This information aided in determining the most efficient compiling parallel configuration for the specific CPU’s available on our cluster, resulting in the fastest VMAT simulation times. Our method was evaluated with calculations involving 10{sup 8} – 10{sup 9} particle histories which are sufficient to verify patient dose using VMAT. Results: Parallelization allowed the calculation of patient dose on the order of 10 – 15 hours with 100 parallel jobs. Due to the compiler optimization process, further speed increases of 23% were achieved when compared with the open-source compiler BEAMnrc packages. Conclusion: Analysis of the BEAMnrc code allowed us to optimize the compiler configuration for VMAT dose calculations. In future work, the optimized MC code, in conjunction with the parallel processing capabilities of BEAMnrc, will be applied to provide accurate and efficient secondary MU checks.

60 APPLIED LIFE SCIENCES↗

p4est : Scalable Algorithms for Parallel Adaptive Mesh Refinement on Forests of Octrees

We present scalable algorithms for parallel adaptive mesh refinement and coarsening (AMR), partitioning, and 2:1 balancing on computational domains composed of multiple connected two-dimensional quadtrees or three-dimensional octrees, referred to as a forest of octrees. By distributing the union of octants from all octrees in parallel, we combine the high scalability proven previously for adaptive single-octree algorithms with the geometric flexibility that can be achieved by arbitrarily connected hexahedral macromeshes, in which each macroelement is the root of an adapted octree. A key concept of our approach is an encoding scheme of the interoctree connectivity that permits arbitrary relative orientations between octrees. Based on this encoding we develop interoctree transformations of octants. These form the basis for high-level parallel octree algorithms, which are designed to interact with an application code such as a numerical solver for partial differential equations. We have implemented and tested these algorithms in the p4est software library. We demonstrate the parallel scalability of p4est on its own and in combination with two geophysics codes. Using p4est we generate and adapt multioctree meshes with up to $5.13\times10^{11}$ octants on as many as 220,320 CPU cores and execute the 2:1 balance algorithm in less than 10 seconds per million octants per process.

Mathematics↗

Manycore Parallel Computing for a Hybridizable Discontinuous Galerkin Nested Multigrid Method

In this work, we present a parallel computing strategy for a hybridizable discontinuous Galerkin (HDG) nested geometric multigrid (GMG) solver. Parallel GMG solvers require a combination of coarse-grain and fine-grain parallelism to improve time-to-solution performance. In this work we focus on fine-grain parallelism. We use Intel's second generation Xeon Phi (Knights Landing) manycore processor. The GMG method achieves ideal convergence rates of 0.2 or less, for high polynomial orders. A matrix free (assembly free) technique is exploited to save considerable memory usage and increase arithmetic intensity. HDG enables static condensation, and due to the discontinuous nature of the discretization, we developed a matrix vector multiply routine that does not require any costly synchronizations or barriers. Our algorithm is able to attain 80% of peak bandwidth performance for higher order polynomials. This is possible due to the data locality inherent in the HDG method. Very high performance is realized for high order schemes, due to good arithmetic intensity, which declines as the order is reduced.

97 MATHEMATICS AND COMPUTING↗

Parallel-In-Time Magnus Integrators

Magnus integrators are a subset of geometric integration methods for the numerical solution of ordinary differential equations that conserve certain invariants in the numerical solution. Our report explores temporal parallelism of Magnus integrators, particularly in the context of nonlinear problems. The method combines the concurrent computation of matrix commutators and exponentials within a time step with a pipelined iteration applied to multiple time steps in parallel. The accuracy and efficiency of time parallel Magnus methods up to order six are highlighted through numerical examples and demonstrate that critical parallel speedup is possible compared to serial methods.

97 MATHEMATICS AND COMPUTING↗

Design and Implementation of “Many Parallel Task” Hybrid Subsurface Model

Continuum scale models have been used to study subsurface flow, transport, and reactions for many years. Recently, pore scale models, which operate at scales of individual soil grains, have been developed to more accurately model pore scale phenomena, such as precipitation, that may not be well represented at the continuum scale. However, particle-based models become prohibitively expensive for modeling realistic domains. Instead, we are developing a hybrid model that simulates the full domain at continuum scale and applies the pore model only to areas of high reactivity. The hybrid model uses a dimension reduction approach to formulate the mathematical exchange of information across scales. Since the location, size, and number of pore regions in the model varies, an adaptive Pore Generator is being implemented to define pore regions at each iteration. A fourth code will provide data transformation from the pore scale back to the continuum scale. These components are coupled into a single hybrid model using the SWIFT workflow system. Our hybrid model workflow simulates a kinetic controlled mixing reaction in which multiple pore-scale simulations occur for every continuum scale timestep. Each pore-scale simulation is itself parallel, thus exhibiting multi-level parallelism. Our workflow manages these multiple parallel tasks simultaneously, with the number of tasks changing across iterations. It also supports dynamic allocation of job resources and visualization processing at each iteration. We discuss the design, implementation and challenges associated with building a scalable, Many Parallel Task, hybrid model to run efficiently on thousands to tens of thousands of processors.

workflow, Many task, Subsurface↗

A Parallel Solver for Graph Laplacians

Problems from graph drawing, spectral clustering, network flow and graph partitioning can all be expressed in terms of graph Laplacian matrices. There are a variety of practical approaches to solving these problems in serial. However, as problem sizes increase and single core speeds stagnate, parallelism is essential to solve such problems quickly. We present an unsmoothed aggregation multigrid method for solving graph Laplacians in a distributed memory setting. We introduce new parallel aggregation and low degree elimination algorithms targeted specifically at irregular degree graphs. These algorithms are expressed in terms of sparse matrix-vector products using generalized sum and product operations. This formulation is amenable to linear algebra using arbitrary distributions and allows us to operate on a 2D sparse matrix distribution, which is necessary for parallel scalability. Our solver outperforms the natural parallel extension of the current state of the art in an algorithmic comparison. We demonstrate scalability to 576 processes and graphs with up to 1.7 billion edges.

Konolige, Tristan↗

TuckerMPI: A Parallel C++/MPI Software Package for Large-scale Data Compression via the Tucker Tensor Decomposition

With this study, our goal is compression of massive-scale grid-structured data, such as the multi-terabyte output of a high-fidelity computational simulation. For such data sets, we have developed a new software package called TuckerMPI, a parallel C++/MPI software package for compressing distributed data. The approach is based on treating the data as a tensor, i.e., a multidimensional array, and computing its truncated Tucker decomposition, a higher-order analogue to the truncated singular value decomposition of a matrix. The result is a low-rank approximation of the original tensor-structured data. Compression efficiency is achieved by detecting latent global structure within the data, which we contrast to most compression methods that are focused on local structure. In this work, we describe TuckerMPI, our implementation of the truncated Tucker decomposition, including details of the data distribution and in-memory layouts, the parallel and serial implementations of the key kernels, and analysis of the storage, communication, and computational costs. We test the software on 4.5 and 6.7 terabyte data sets distributed across 100 s of nodes (1,000 s of MPI processes), achieving compression ratios between 100 and 200,000×, which equates to 99--99.999% compression (depending on the desired accuracy) in substantially less time than it would take to even read the same dataset from a parallel file system. Moreover, we show that our method also allows for reconstruction of partial or down-sampled data on a single node, without a parallel computer so long as the reconstructed portion is small enough to fit on a single machine, e.g., in the instance of reconstructing/visualizing a single down-sampled time step or computing summary statistics. The code is available at https://gitlab.com/tensors/TuckerMPI.

97 MATHEMATICS AND COMPUTING↗

Generating Massive Scale-free Networks: Novel Parallel Algorithms using the Preferential Attachment Model

Recently, there has been substantial interest in the study of various random networks as mathematical models of complex systems. As real-life complex systems grow larger, the ability to generate progressively large random networks becomes all the more important. This motivates the need for efficient parallel algorithms for generating such networks. Naïve parallelization of sequential algorithms for generating random networks is inefficient due to inherent dependencies among the edges and the possibility of creating duplicate (parallel) edges. In this article, we present message passing interface-based distributed memory parallel algorithms for generating random scale-free networks using the preferential-attachment model. Our algorithms are experimentally verified to scale very well to a large number of processing elements (PEs), providing near-linear speedups. The algorithms have been exercised with regard to scale and speed to generate scale-free networks with one trillion edges in 6 minutes using 1,000 PEs.

97 MATHEMATICS AND COMPUTING↗

PLANC: Parallel Low-rank Approximation with Nonnegativity Constraints

In this work, we consider the problem of low-rank approximation of massive dense nonnegative tensor data, for example, to discover latent patterns in video and imaging applications. As the size of data sets grows, single workstations are hitting bottlenecks in both computation time and available memory. We propose a distributed-memory parallel computing solution to handle massive data sets, loading the input data across the memories of multiple nodes, and performing efficient and scalable parallel algorithms to compute the low-rank approximation. We present a software package called Parallel Low-rank Approximation with Nonnegativity Constraints, which implements our solution and allows for extension in terms of data (dense or sparse, matrices or tensors of any order), algorithm (e.g., from multiplicative updating techniques to alternating direction method of multipliers), and architecture (we exploit GPUs to accelerate the computation in this work). We describe our parallel distributions and algorithms, which are careful to avoid unnecessary communication and computation, show how to extend the software to include new algorithms and/or constraints, and report efficiency and scalability results for both synthetic and real-world data sets.

97 MATHEMATICS AND COMPUTING↗

Lowering and Runtime Support for Fortran’s Multi-Image Parallel Features using LLVM Flang, PRIF, and Caffeine

This paper provides an overview of the multi-image parallel features in Fortran 2023 and their implementation in the LLVM flang compiler and the Caffeine parallel runtime library. The features of interest support a Single-Program, Multiple-Data (SPMD) programming model based on executing multiple “images”, each of which is a program instance. The features also support a Partitioned Global Address Space (PGAS) in the form of “coarray” distributed data structures. The paper discusses the lowering of multi-image features to the Parallel Runtime Interface for Fortran (PRIF) and the implementation of PRIF in the Caffeine parallel runtime library. This paper also provides an early view into the design of a new multi-image dialect of the LLVM Multi-Level Intermediate Representation (MLIR). We describe validation and testing of the resulting software stack, and demonstrate that performance compares favorably to another open-source compiler and runtime library: GNU Compiler Collection (GCC) gfortran and OpenCoarrays, respectively.

Bonachea, Dan↗

Parallel k-means++

A parallelization of the k-means++ seed selection algorithm on three distinct hardware platforms: GPU, multicore CPU, and multithreaded architecture. K-means++ was developed by David Arthur and Sergei Vassilvitskii in 2007 as an extension of the k-means data clustering technique. These algorithms allow people to cluster multidimensional data, by attempting to minimize the mean distance of data points within a cluster. K-means++ improved upon traditional k-means by using a more intelligent approach to selecting the initial seeds for the clustering process. While k-means++ has become a popular alternative to traditional k-means clustering, little work has been done to parallelize this technique. We have developed original C++ code for parallelizing the algorithm on three unique hardware architectures: GPU using NVidia's CUDA/Thrust framework, multicore CPU using OpenMP, and the Cray XMT multithreaded architecture. By parallelizing the process for these platforms, we are able to perform k-means++ clustering much more quickly than it could be done before.

Source record↗