SEARCH · Search NASA

Results for “Parallel Optimization Algorithms”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3

All sources NASA NTRS DOE OSTI

NASA NTRS2012

Recursive Branching Simulated Annealing Algorithm

This innovation is a variation of a simulated-annealing optimization algorithm that uses a recursive-branching structure to parallelize the search of a parameter space for the globally optimal solution to an objective. The algorithm has been demonstrated to be more effective at searching a parameter space than traditional simulated-annealing methods for a particular problem of interest, and it can readily be applied to a wide variety of optimization problems, including those with a parameter space having both discrete-value parameters (combinatorial) and continuous-variable parameters. It can take the place of a conventional simulated- annealing, Monte-Carlo, or random- walk algorithm. In a conventional simulated-annealing (SA) algorithm, a starting configuration is randomly selected within the parameter space. The algorithm randomly selects another configuration from the parameter space and evaluates the objective function for that configuration. If the objective function value is better than the previous value, the new configuration is adopted as the new point of interest in the parameter space. If the objective function value is worse than the previous value, the new configuration may be adopted, with a probability determined by a temperature parameter, used in analogy to annealing in metals. As the optimization continues, the region of the parameter space from which new configurations can be selected shrinks, and in conjunction with lowering the annealing temperature (and thus lowering the probability for adopting configurations in parameter space with worse objective functions), the algorithm can converge on the globally optimal configuration. The Recursive Branching Simulated Annealing (RBSA) algorithm shares some features with the SA algorithm, notably including the basic principles that a starting configuration is randomly selected from within the parameter space, the algorithm tests other configurations with the goal of finding the globally optimal solution, and the region from which new configurations can be selected shrinks as the search continues. The key difference between these algorithms is that in the SA algorithm, a single path, or trajectory, is taken in parameter space, from the starting point to the globally optimal solution, while in the RBSA algorithm, many trajectories are taken; by exploring multiple regions of the parameter space simultaneously, the algorithm has been shown to converge on the globally optimal solution about an order of magnitude faster than when using conventional algorithms. Novel features of the RBSA algorithm include: 1. More efficient searching of the parameter space due to the branching structure, in which multiple random configurations are generated and multiple promising regions of the parameter space are explored; 2. The implementation of a trust region for each parameter in the parameter space, which provides a natural way of enforcing upper- and lower-bound constraints on the parameters; and 3. The optional use of a constrained gradient- search optimization, performed on the continuous variables around each branch s configuration in parameter space to improve search efficiency by allowing for fast fine-tuning of the continuous variables within the trust region at that configuration point.

Bolcar, Matthew↗

DOE OSTI2022

Optimizing the hit finding algorithm for liquid argon TPC neutrino detectors using parallel architectures

Neutrinos are particles that interact rarely, so identifying them requires large detectors which produce lots of data. Processing this data with the computing power available is becoming even more difficult as the detectors increase in size to reach their physics goals. Liquid argon time projection chamber (LArTPC) neutrino experiments are expected to grow in the next decade to have 100 times more wires than in currently operating experiments, and modernization of LArTPC reconstruction code, including parallelization both at data- and instruction-level, will help to mitigate this challenge. The LArTPC hit finding algorithm is used across multiple experiments through a common software framework. In this paper we discuss a parallel implementation of this algorithm. Using a standalone setup we find speedup factors of two times from vectorization and 30–100 times from multi-threading on Intel architectures. The new version has been incorporated back into the framework so that it can be used by experiments. On a serial execution, the integrated version is about 10 times faster than the previous one and, once parallelization is enabled, more speedups comparable to the standalone program are achieved.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2018

Parallel Nonnegative CP Decomposition of Dense Tensors

The CP tensor decomposition is a low-rank approximation of a tensor. We present a distributed-memory parallel algorithm and implementation of an alternating optimization method for computing a CP decomposition of dense tensors that can enforce nonnegativity of the computed low-rank factors. The principal task is to parallelize the Matricized-Tensor Times Khatri-Rao Product (MTTKRP) bottleneck subcomputation. The algorithm is computation efficient, using dimension trees to avoid redundant computation across MTTKRPs within the alternating method. Our approach is also communication efficient, using a data distribution and parallel algorithm across a multidimensional processor grid that can be tuned to minimize communication. We benchmark our software on synthetic as well as hyperspectral image and neuroscience dynamic functional connectivity data, demonstrating that our algorithm scales well to 100s of nodes (up to 4096 cores) and is faster and more general than the currently available parallel software.

Ballard, Grey↗

NASA NTRS2000

High Performance Fortran for Aerospace Applications

This paper focuses on the use of High Performance Fortran (HPF) for important classes of algorithms employed in aerospace applications. HPF is a set of Fortran extensions designed to provide users with a high-level interface for programming data parallel scientific applications, while delegating to the compiler/runtime system the task of generating explicitly parallel message-passing programs. We begin by providing a short overview of the HPF language. This is followed by a detailed discussion of the efficient use of HPF for applications involving multiple structured grids such as multiblock and adaptive mesh refinement (AMR) codes as well as unstructured grid codes. We focus on the data structures and computational structures used in these codes and on the high-level strategies that can be expressed in HPF to optimally exploit the parallelism in these algorithms.

Mehrotra, Piyush↗

DOE OSTI2019

miniVite: A Graph Analytics Benchmarking Tool for Massively Parallel Systems

Benchmarking of high performance computing sys-tems can help provide critical insights for ef?cient design of computing systems and software applications. Although a large number of tools for benchmarking exist, there is a lack of rep-resentative benchmarks for the class of irregular computations as exempli?ed by graph analytics. In this paper, we propose miniVite as a representative graph analytics benchmark tool to test a variety of distributed-memory systems. Graph clustering, popularly known as community detection, is a prototypical graph operation used in numerous scienti?c computing and analytics applications. The goal of clustering is to partition a graph into clusters (or communities) such that each cluster consists of vertices that are densely connected within the cluster and sparsely connected to the rest of the graph. Modu-larity optimization is a popular technique for identifying clusters in a graph. Ef?cient parallelization of modularity optimization-based algorithms is challenging. One successful approach was conceived in Vite, a distributed-memory implementation of the Louvain algorithm that incorporates several heuristics. We introduce miniVite as a representative but simpli?ed variant of Vite, to serve as a prototypical graph analytics benchmarking tool. Unlike other graph-based methods such as breadth-?rst search and betweenness centrality, miniVite represents highly complex computational patterns stressing a variety of system features, which can provide crucial insight for co-design of future computing systems.

community detection, parallel, MPI↗

DOE OSTI2023

Parallel Memory-Independent Communication Bounds for SYRK

In this paper, we focus on the parallel communication cost of multiplying a matrix with its transpose, known as a symmetric rank-k update (SYRK). SYRK requires half the computation of general matrix multiplication because of the symmetry of the output matrix. Recent work (Beaumont et al., SPAA '22) has demonstrated that the sequential I/O complexity of SYRK is also a constant factor smaller than that of general matrix multiplication. Inspired by this progress, we establish memory-independent parallel communication lower bounds for SYRK with smaller constants than general matrix multiplication, and we show that these constants are tight by presenting communication-optimal algorithms. The crux of the lower bound proof relies on extending a key geometric inequality to symmetric computations and analytically solving a constrained nonlinear optimization problem. Here, the optimal algorithms use a triangular blocking scheme for parallel distribution of the symmetric output matrix and corresponding computation.

Communication costs↗

NASA NTRS2019

High-Performance Computing Optimization for Aladyn – Adaptive Neural Network Molecular Dynamics Mini-Application

This report provides a description and performance evaluation of the optimization techniques for high performance computing (HPC) implementation of the open source Computational Materials mini-application Aladyn (https://github.com/nasa/aladyn). Aladyn is a basic molecular dynamics code written in FORTRAN 2003, which is designed to demonstrate the use of adaptive neural networks (ANNs) in atomistic simulations. The role of ANNs is to efficiently reproduce the very complex energy landscape resulting from the atomic interactions in materials with the accuracy of the more expensive quantum mechanics-based calculations. The ANN is trained on a large set of atomic structures calculated using the density functional theory (DFT) method. While achieving orders of magnitude faster computational performance than DFT, the ANN-based approach was still very computationally demanding compared to the conventional approach of using empirically fitted energy functions. After its initial development, Aladyn was evaluated and optimized by experts at the NASA Advanced Supercomputing (NAS) division to exploit modern supercomputer architectures. The code has been optimized for execution on multicore central processing units (CPUs), including Intel® Skylake microarchitecture, and on graphic accelerators, such as Nvidia® V100 graphic processing units (GPUs), using Open Multi-Processing (OpenMP) and Open Accelerators (OpenACC) programming interfaces. The optimization achieved a speedup of 4.7 times the baseline version on CPU performance and an additional 2.4 times on CPU+GPU performance. Atomistic computer simulations are a fundamental tool in materials research to model material properties form physics-based first principles. Atomic interaction, governed by Quantum Mechanics (QM) require sophisticated and highly computationally demanding mathematical models to calculate [1]. Classical methods use approximate functional forms, empirically fitted through a set of variable parameters to emulate atomic energies as direct functions of atomic coordinates [2]. While empirical potentials are computationally much simpler, allowing simulations of large-scale systems of up to a trillion (1012) atoms [3], they are substantially less accurate compared to quantum calculations and applicable only to very specific atomic configurations or predefined crystallographic phases. A recently suggested approach is to use heuristic machine learning methods [4], such as those based on Adaptive Neural Networks (ANNs) to predict atomic energies, after being trained on a sufficiently large database of QM-calculated structures [5,6]. This approach reduces significantly the computational complexity, allowing for simulations of orders of magnitude larger systems compared to QM-based methods without compromising accuracy. Still, compared to classical methods using empirical energy functions, ANN methods remain two- to three orders of magnitude more computationally demanding. Hence, the computational cost of simulations, together with the need for extensive training of ANNs, still makes the practical implementation of ANN-based methods quite challenging. The purpose of the Aladyn mini-application software [7], available as open source at https://github.com/nasa/aladyn, is to be a testbed for exploring possible optimization strategies to develop highly scalable parallel algorithms for ANN-based atomistic simulations. Aladyn is aimed at utilizing the architecture of the high-end modern highperformance computing (HPC) hardware based on multicore central processing units (CPUs) equipped with graphic processing unit (GPU) accelerators. Specifically, the goal is to optimize the performance on a single HPC compute node, before implementing scaling to multi-node parallelization using message passing interface (MPI). At the same time, the open source code of Aladyn can serve as a training model for students and professors in academia.

Yamakov, Vesselin I.↗

DOE OSTI2020

GentenMPI: Distributed Memory Sparse Tensor Decomposition

GentenMPl is a toolkit of sparse canonical polyadic (CP) tensor decomposition algorithms that is designed to run effectively on distributed-memory high-performance computers. Its use of distributed-memory parallelism enables it to efficiently decompose tensors that are too large for a single compute node's memory. GentenMPl leverages Sandia's decades-long investment in the Trilinos solver framework for much of its parallel-computation capability. Trilinos contains numerical algorithms and linear algebra classes that have been optimized for parallel simulation of complex physical phenomena. This work applies these tools to the data science problem of sparse tensor decomposition. In this report, we describe the use of Trilinos in GentenMPl, extensions needed for sparse tensor decomposition, and implementations of the CP-ALS (CP via alternating least squares) and GCP-SGD (generalized CP via stochastic gradient descent) sparse tensor decomposition algorithms. We show that GentenMPl can decompose sparse tensors of extreme size, e.g., a 12.6-terabyte tensor on 8192 computer cores. We demonstrate that the Trilinos backbone provides good strong and weak scaling of the tensor decomposition algorithms.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1998

Variable-Complexity Multidisciplinary Optimization on Parallel Computers

This report covers work conducted under grant NAG1-1562 for the NASA High Performance Computing and Communications Program (HPCCP) from December 7, 1993, to December 31, 1997. The objective of the research was to develop new multidisciplinary design optimization (MDO) techniques which exploit parallel computing to reduce the computational burden of aircraft MDO. The design of the High-Speed Civil Transport (HSCT) air-craft was selected as a test case to demonstrate the utility of our MDO methods. The three major tasks of this research grant included: development of parallel multipoint approximation methods for the aerodynamic design of the HSCT, use of parallel multipoint approximation methods for structural optimization of the HSCT, mathematical and algorithmic development including support in the integration of parallel computation for items (1) and (2). These tasks have been accomplished with the development of a response surface methodology that incorporates multi-fidelity models. For the aerodynamic design we were able to optimize with up to 20 design variables using hundreds of expensive Euler analyses together with thousands of inexpensive linear theory simulations. We have thereby demonstrated the application of CFD to a large aerodynamic design problem. For the predicting structural weight we were able to combine hundreds of structural optimizations of refined finite element models with thousands of optimizations based on coarse models. Computations have been carried out on the Intel Paragon with up to 128 nodes. The parallel computation allowed us to perform combined aerodynamic-structural optimization using state of the art models of a complex aircraft configurations.

Grossman, Bernard↗

NASA NTRS1990

Assignment Of Finite Elements To Parallel Processors

Elements assigned approximately optimally to subdomains. Mapping algorithm based on simulated-annealing concept used to minimize approximate time required to perform finite-element computation on hypercube computer or other network of parallel data processors. Mapping algorithm needed when shape of domain complicated or otherwise not obvious what allocation of elements to subdomains minimizes cost of computation.

Salama, Moktar A.↗

NASA NTRS2007

Efficient Optimization of Low-Thrust Spacecraft Trajectories

A paper describes a computationally efficient method of optimizing trajectories of spacecraft driven by propulsion systems that generate low thrusts and, hence, must be operated for long times. A common goal in trajectory-optimization problems is to find minimum-time, minimum-fuel, or Pareto-optimal trajectories (here, Pareto-optimality signifies that no other solutions are superior with respect to both flight time and fuel consumption). The present method utilizes genetic and simulated-annealing algorithms to search for globally Pareto-optimal solutions. These algorithms are implemented in parallel form to reduce computation time. These algorithms are coupled with either of two traditional trajectory- design approaches called "direct" and "indirect." In the direct approach, thrust control is discretized in either arc time or arc length, and the resulting discrete thrust vectors are optimized. The indirect approach involves the primer-vector theory (introduced in 1963), in which the thrust control problem is transformed into a co-state control problem and the initial values of the co-state vector are optimized. In application to two example orbit-transfer problems, this method was found to generate solutions comparable to those of other state-of-the-art trajectory-optimization methods while requiring much less computation time.

Lee, Seungwon↗

DOE OSTI2019

DECA: scalable XHMM exome copy-number variant calling with ADAM and Apache Spark

Abstract Background XHMM is a widely used tool for copy-number variant (CNV) discovery from whole exome sequencing data but can require hours to days to run for large cohorts. A more scalable implementation would reduce the need for specialized computational resources and enable increased exploration of the configuration parameter space to obtain the best possible results. Results DECA is a horizontally scalable implementation of the XHMM algorithm using the ADAM framework and Apache Spark that incorporates novel algorithmic optimizations to eliminate unneeded computation. DECA parallelizes XHMM on both multi-core shared memory computers and large shared-nothing Spark clusters. We performed CNV discovery from the read-depth matrix in 2535 exomes in 9.3 min on a 16-core workstation (35.3× speedup vs. XHMM), 12.7 min using 10 executor cores on a Spark cluster (18.8× speedup vs. XHMM), and 9.8 min using 32 executor cores on Amazon AWS’ Elastic MapReduce. We performed CNV discovery from the original BAM files in 292 min using 640 executor cores on a Spark cluster. Conclusions We describe DECA’s performance, our algorithmic and implementation enhancements to XHMM to obtain that performance, and our lessons learned porting a complex genome analysis application to ADAM and Spark. ADAM and Apache Spark are a performant and productive platform for implementing large-scale genome analyses, but efficiently utilizing large clusters can require algorithmic optimizations and careful attention to Spark’s configuration parameters.

Linderman, Michael D. (ORCID:0000000296437148)↗

NASA NTRS2018

Solution of Ordinary Differential Equations in Gradient-Based Multidisciplinary Design Optimization

A gradient-based approach to multidisciplinary design optimization enables efficient scalability to large numbers of design variables. However, the need for derivatives causes difficulties when integrating ordinary differential equations (ODEs) in models. To simplify this, we propose the use of the general linear methods framework, which unifies all Runge-Kutta and linear multistep methods. This approach enables rapid implementation of integration methods without the need to differentiate each one, even in a gradient-based optimization context. We also develop a new parallel time integration algorithm that enables vectorization across time steps. We present a set of benchmarking results using a stiff ODE, a non-stiff nonlinear ODE, and an orbital dynamics ODE, and compare integration methods. In a modular gradient-based multidisciplinary design optimization context, we find that the new parallel time integration algorithm with high-order implicit methods, especially Gauss-Legendre collocation, is the best choice for a broad range of problems.

Hwang, John T.↗

DOE OSTI2018

Optimization of groundwater artificial recharge systems using a genetic algorithm: a case study in Beijing, China

An optimization approach is used for the operation of groundwater artificial recharge systems in an alluvial fan in Beijing, China. The optimization model incorporates a transient groundwater flow model, which allows for simulation of the groundwater response to artificial recharge. The facilities’ operation with regard to recharge rates is formulated as a nonlinear programming problem to maximize the volume of surface water recharged into the aquifers under specific constraints. This optimization problem is solved by the parallel genetic algorithm (PGA) based on OpenMP, which could substantially reduce the computation time. To solve the PGA with constraints, the multiplicative penalty method is applied. In addition, the facilities’ locations are implicitly determined on the basis of the results of the recharge-rate optimizations. Two scenarios are optimized and the optimal results indicate that the amount of water recharged into the aquifers will increase without exceeding the upper limits of the groundwater levels. Optimal operation of this artificial recharge system can also contribute to the more effective recovery of the groundwater storage capacity.

58 GEOSCIENCES↗

NASA NTRS1998

Massively Parallel Solution of Poisson Equation on Coarse Grain MIMD Architectures

In this paper a new algorithm, designated as Fast Invariant Imbedding algorithm, for solution of Poisson equation on vector and massively parallel MIMD architectures is presented. This algorithm achieves the same optimal computational efficiency as other Fast Poisson solvers while offering a much better structure for vector and parallel implementation. Our implementation on the Intel Delta and Paragon shows that a speedup of over two orders of magnitude can be achieved even for moderate size problems.

parallel algorithms MIMD parallel architectures↗

DOE OSTI2023

PyFLEXTRKR: a flexible feature tracking Python software for convective cloud analysis

Abstract. This paper describes the new open-source framework PyFLEXTRKR (Python FLEXible object TRacKeR), a flexible atmospheric feature tracking software package with specific capabilities to track convective clouds from a variety of observations and model simulations. This software can track any atmospheric 2D objects and handle merging and splitting explicitly. The package has a collection of multi-object identification algorithms, scalable parallelization options, and has been optimized for large datasets including global high-resolution data. We demonstrate applications of PyFLEXTRKR on tracking individual deep convective cells and mesoscale convective systems from observations and model simulations ranging from large-eddy resolving (∼100s m) to mesoscale (∼10s km) resolutions. Visualization, post-processing, and statistical analysis tools are included in the package. New Lagrangian analyses of convective clouds produced by PyFLEXTRKR applicable to a wide range of datasets and scales facilitate advanced model evaluation and development efforts as well as scientific discovery.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Progress on Optimizing Wind Farms and Rotor Designs Using Adjoints

Modern wind plants are increasingly tasked with multiple performance objectives. In addition to designing plants that maximize power output and minimize the levelized cost of energy (LCOE), the design and operation of wind plants is increasingly influenced by challenges regarding grid integration of variable generation renewables. This places a growing emphasis on making wind plants more controllable and predictable. WindSE is a Reynolds-averaged Navier-Stokes (RANS) model designed around analytical gradient and adjoint methods, with the ability to capture terrain-induced effects, as shown in Figure 1. The recent addition of an unsteady solver with an actuator line method (ALM) and ongoing work to enable massively parallel optimizations gives it a unique niche to explore coupled plant-level controls and design problems. This code is an open source python package built on the FEniCS framework that utilizes fast, parallel PETSc solvers to model fluid flow throughout wind-farm scale domains. Two recent studies performed using WindSE demonstrate the capability to optimize under a wide variety of flow conditions and objective functions. In the first, we present an optimization focused on modifying the layout of a wind farm with a fixed number of turbines for maximum total power output [1]. This study highlights the ability to quickly perform simulations using the steady Navier-Stokes solver combined with rotors represented as actuator disks while also stressing the importance of capturing terrain-induced effects. Gradient-based optimization using the RANS equations is viable due to the inclusion of efficiently computed adjoint derivatives. We interpret the physical results of the optimal layout and also discuss the computational cost of scaling to larger problems. In the second study, we present the capabilities of the unsteady Navier-Stokes solver, where rotor-blade profiles represented by actuator lines are optimized to enhance wake steering effects and overall power production [2]. We quantify the wind plant performance gains obtained from this type of simultaneous control co-design optimization as compared to optimizing the blade design and yaw independently. Figure 2 shows the differences between a baseline two-turbine system and an optimized system where we fine-tune the blade chord profile. Results and challenges from each study are quickly summarized and used to motivate the current development efforts within WindSE. Current and future work is focused on enabling higher-resolution studies with more degrees of freedom through parallelization of both the simulation and optimization algorithms. We present benchmarking results to show that WindSE performs well in both weak- and strong-scaling tests and further demonstrate that the optimizer obtains the same convergence rates in both shared- and distributed-memory environments. Using larger wind farms, we can study deep-array effects within an optimization context, allowing the use of objective functions that have been previously unstudied. As an example, we present ongoing work on a blockage metric which characterizes the loss of available kinetic energy due to wake effects from multiple upstream turbines.

adjoint optimization↗

DOE OSTI2019

HyperSpace

Machine learning (ML) models often contain numerous hyperparameters, free parameters that must be set before the models can be trained. As the number of model hyperparameters increases, their optimization becomes significantly more challenging as we face a combinatorial increase in potential model configurations. Similarly, there is an increased chance that our models’ hyperparameters interact in complex ways. HyperSpace allows a user to optimize complex machine learning algorithms based on their hyperparameters. HyperSpace works by parallelizing parameter search spaces, running Bayesian model based optimization (SMBO) over each of these spaces in parallel. It was designed to be as minimally invasive as possible such that very little change to existing code will be needed to get a user started.

Ramanathan, Arvind↗