Search NASA⌕ Search

SEARCH · Search NASA

Results for “Computational optimization”

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 703 records · Page 39

The SODA Approach: Leveraging High-Level Synthesis for Hardware/Software Co-design and Hardware Specialization: Invited

Novel "converged" applications combine phases of scientific simulation with data analysis and machine learning. Each computational phase can benefit from specialized accelerators. However, algorithms evolve so quickly that mapping them on existing accelerators is suboptimal or even impossible. This paper presents the SODA (Software Defined Accelerators) framework, a modular, multi-level, open-source, no-human-in-the-loop, hardware synthesizer that enables end-to-end generation of specialized accelerators. SODA is composed of SODA-Opt, a high-level frontend developed in MLIR that interfaces with domain-specific programming frameworks and allows performing system level design, and Bambu, a state-of-the-art high-level synthesis engine that can target different device technologies. The framework implements design space exploration as compiler optimization passes. We show how the modular, yet tight, integration of the high-level optimizer and lower-level HLS tools enables the generation of accelerators optimized for the computational patterns of converged applications. We then discuss some of the research opportunities that such a framework allows, including system-level design, profile driven optimization, and supporting new optimization metrics.

Bohm Agostini, Nicolas↗

An Integrated High-performance Computing and Digital Real-time Simulation Testbed to Benchmark Closed-loop Load Shedding Algorithms in Power Systems

An integrated testbed using digital real-time simulator (DRTS) and a high-performance computing (HPC) cluster is presented here to compare speed and performance of computational schemes to mitigate time-critical issues in electric power systems. The first approach in this testbed validation is taken by running a set of closed-loop load shedding algorithms to compare and contrast two paradigms of arresting cascading failure propagation. Two algorithms involve solving DC and AC power flow model-based optimization problems to compute load shedding at different buses, while a model-based stochastic search using parallel computing provides a viable alternative. The algorithms are implemented in the DRTS-HPC testbed for the IEEE 14-bus benchmark transmission system. As a proof of the concept, simulation results are presented for implementation of closed-loop load-shedding algorithms for cascading failures in the DRTS-HPC testbed

24 POWER TRANSMISSION AND DISTRIBUTION↗

Evaluation of stochastic algorithms for financial mathematics problems from point of view of energy-efficiency

The recent developments in the area of high-performance computing are driven not only by the desire for ever higher performance but also by the rising costs of electricity. The use of various types of accelerators like GPUs, Intel Xeon Phi has become mainstream and many algorithms and applications have been ported to make use of them where available. In Financial Mathematics the question of optimal use of computational resources should also take into account the limitations on space, because in many use cases the servers are deployed close to the exchanges. In this work we evaluate various algorithms for option pricing that we have implemented for different target architectures in terms of their energy and space efficiency. Since it has been established that low-discrepancy sequences may be better than pseudorandom numbers for these types of algorithms, we also test the Sobol and Halton sequences. We present the raw results, the computed metrics and conclusions from our tests.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Multi-objective optimization of root phenotypes for nutrient capture using evolutionary algorithms

Root phenotypes are avenues to the development of crop cultivars with improved nutrient capture, which is an important goal for global agriculture. The fitness landscape of root phenotypes is highly complex and multidimensional. It is difficult to predict which combinations of traits (phene states) will create the best performing integrated phenotypes in various environments. Brute force methods to map the fitness landscape by simulating millions of phenotypes in multiple environments are computationally challenging. Evolutionary optimization algorithms may provide more efficient avenues to explore high dimensional domains such as the root phenotypic space. We coupled the three-dimensional functional–structural plant model, SimRoot, to the Borg Multi-Objective Evolutionary Algorithm (MOEA) and the evolutionary search over several generations facilitated the identification of optimal root phenotypes balancing trade-offs across nutrient uptake, biomass accumulation, and root carbon costs in environments varying in nutrient availability. Our results show that several combinations of root phenes generate optimal integrated phenotypes where performance in one objective comes at the cost of reduced performance in one or more of the remaining objectives, and such combinations differed for mobile and non-mobile nutrients and for maize (a monocot) and bean (a dicot). Functional–structural plant models can be used with multi-objective optimization to identify optimal root phenotypes under various environments, including future climate scenarios, which will be useful in developing the more resilient, efficient crops urgently needed in global agriculture.

59 BASIC BIOLOGICAL SCIENCES↗

Efficient frequency allocation for superconducting quantum processors using improved optimization techniques

Building on previous research on frequency allocation optimization for superconducting circuit quantum processors, this work incorporates several techniques to improve overall solution quality. Here, we introduce constraints and imposed edgewise differences help to improve the optimization results. We also introduce optimization variables for the orientation of each edge, defined as the direction from the control qubit to the target qubit, to be chosen during optimization. To scale up to larger processors, multimodule designs are employed with various boundary conditions, thereby enhancing the collective yield. These enhancements allow for greater flexibility in processor design by eliminating the need for handpicked orientations. We support the efficient assembly of large processors with dense connectivity by choosing the best boundary conditions. Examples demonstrate that, at low computational cost, this optimization approach finds a frequency configuration for a square chip with over 1000 qubits and over 10% yield at much larger dispersion levels than required by previous approaches.

Zhang, Zewen [Argonne National Laboratory (ANL), A↗

Resilience-based Recovery Scheduling of Transportation Network in Mixed Traffic Environment: A Deep-Ensemble-Assisted Active Learning Approach

Devising effective post-hazard recovery strategies is critical in enhancing the resilience of transportation networks (TNs). However, existing work does not consider the multiclass users’ travel behavior in network functionality quantification and the metaheuristic solution procedures often suffer from extensive computational burden due to the exploration need in large solution space and the expensive functionality quantification. Here, this study develops a bilevel decision-making framework for the resilience-based recovery scheduling of the TN in a mixed traffic environment with connected and autonomous vehicles (CAVs) and human-driven vehicles (HDVs). The lower level quantifies the TN's functionality over time considering different travel behavior of CAV and HDV users arisen from their different levels of information perception. The upper level presents a novel deep-ensemble-assisted active learning approach to balance optimization performance and computational cost. This framework can help decision makers better quantify the TN's functionality to support effective recovery scheduling of TN with different mixed traffic scenarios ranging from HDV-only to future CAV-dominant traffic. The optimization approach bears the potential to be extended to solving general large-scale network recovery scheduling problems effectively and efficiently. The proposed methodology is demonstrated using a real-world traffic network in Southern California under earthquake considering deterministic and stochastic repair durations.

42 ENGINEERING↗

Short-depth QAOA circuits and quantum annealing on higher-order ising models

Abstract We present a direct comparison between QAOA (Quantum Alternating Operator Ansatz), and QA (Quantum Annealing) on 127 qubit problem instances. QAOA with p = 1, 2 rounds is executed on the 127 qubit heavy-hex graph gate-model quantum computer ibm_washington, using on-device grid-searches for angle finding, and QA is executed on two Pegasus-chip D-Wave quantum annealers. The problems are random Ising models whose connectivity matches heavy-hex graphs and the Pegasus graph connectivity, and optionally include hardware-compatible cubic terms ( Z Z Z terms). The QAOA circuits are heavily optimized and of extremely short depth, with a CNOT depth of 6 per round, which allows whole chip usage of the heavy-hex lattice. QAOA and QA are both compared against simulated annealing and the optimal solutions are computed exactly using CPLEX. The noiseless mean QAOA expectation values for p = 1, 2 are computed using classical light-cone based simulations. We find QA outperforms QAOA on the evaluated devices.

127 qubits↗

Fuzzified PaCcET for Economic-Emission Scheduling of Microgrids

In this paper, a new approach is proposed to solve a multi-objective economic-emission scheduling problem in microgrids (MGs) by simultaneously minimizing the energy and emission costs of the MG with various distributed energy resources (DERs). The proposed approach is an extension of a computationally effective multiobjective optimization technique, Pareto concavity elimination transformation (PaCcET). The proposed approach, referred to as Fuzzified-PaCcET, employs a fuzzy logic controller to dynamically revise crossover and mutation rates in the original PaCcET leading to the faster convergence of the solution. The proposed approach finds the best Pareto front, also referred to as a Non-dominated set (NDS) of solutions, instead of finding a single optimal solution. In order to find the solutions on concave areas of the Pareto front, an iterative objective space transformation is performed in the PaCcET algorithm to allow a linear combination of objective functions (in the transformed objective space). The proposed Fuzzified-PaCcET-based scheduling is implemented on a MG with various dispatchable and non-dispatchable DERs to find the set of optimal solutions according to the total fuel cost of DERs, as well as the most optimum environmental cost. In order to extract the best compromise solution (BCS) among NDS of solutions, a fuzzy-based method is implemented. The comparison of the simulation results of the Fuzzified-PaCcET with that of PaCcET shows that Fuzzified-PaCcET can generate better solution with less computational burden.

Gautam, Mukesh↗

Optimizing Interacting Potentials to Form Targeted Materials Structures

Conventional applications of the principles of statistical mechanics (the "forward" problems), start with particle interaction potentials, and proceed to deduce local structure and macroscopic properties. Other applications (that may be classified as "inverse" problems), begin with targeted configurational information, such as low-order correlation functions that characterize local particle order, and attempt to back out full-system configurations and/or interaction potentials. To supplement these successful experimental and numerical "forward" approaches, we have focused on inverse approaches that make use of analytical and computational tools to optimize interactions for targeted self-assembly of nanosystems. The most original aspect of our work is its inherently inverse approach: instead of predicting structures that result from given interaction potentials among particles, we determine the optimal potential that most robustly stabilizes a given target structure subject to certain constraints. Our inverse approach could revolutionize the manner in which materials are designed and fabricated. There are a number of very tangible properties (e.g. zero thermal expansion behavior), elastic constants, optical properties for photonic applications, and transport properties.

36 MATERIALS SCIENCE↗

Machine learning-based surrogate modeling for data-driven optimization: a comparison of subset selection for regression techniques

Optimization of simulation-based or data-driven systems is a challenging task, which has attracted significant attention in the recent literature. A very efficient approach for optimizing systems without analytical expressions is through fitting surrogate models. Due to their increased flexibility, nonlinear interpolating functions, such as radial basis functions and Kriging, have been predominantly used as surrogates for data-driven optimization; however, these methods lead to complex nonconvex formulations. Alternatively, commonly used regression-based surrogates lead to simpler formulations, but they are less flexible and inaccurate if the form is not known a priori. In this work, we investigate the efficiency of subset selection regression techniques for developing surrogate functions that balance both accuracy and complexity. Subset selection creates sparse regression models by selecting only a subset of original features, which are linearly combined to generate a diverse set of surrogate models. Five different subset selection techniques are compared with commonly used nonlinear interpolating surrogate functions with respect to optimization solution accuracy, computation time, sampling requirements, and model sparsity. Furthermore, our results indicate that subset selection-based regression functions exhibit promising performance when the dimensionality is low, while interpolation performs better for higher dimensional problems.

42 ENGINEERING↗

Mr. Scan: extreme scale density-based clustering using a tree-based network of GPGPU nodes, In: SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

Density-based clustering algorithms are a widely-used class of data mining techniques that can find irregularly shaped clusters and cluster data without prior knowledge of the number of clusters it contains. DBSCAN is the most well-known density-based clustering algorithm. We introduce our version of DBSCAN, called Mr. Scan, which uses a hybrid parallel implementation that combines the MRNet tree-based distribution network with GPGPU-equipped nodes. Mr. Scan avoids the problems of existing implementations by effectively partitioning the point space and by optimizing DBSCAN's computation over dense data regions. We tested Mr. Scan on both a geolocated Twitter dataset and image data obtained from the Sloan Digital Sky Survey. At its largest scale, Mr. Scan clustered 6.5 billion points from the Twitter dataset on 8,192 GPU nodes on Cray Titan in 17.3 minutes. All other parallel DBSCAN implementations have only demonstrated the ability to cluster up to 100 million points.

Computer Science↗

Computational Spectroscopy of the Cr–Cr Bond in Coordination Complexes

In this work, we report the accurate computational vibrational analysis of the Cr–Cr bond in dichromium complexes using second-order multireference complete active space methods (CASPT2), allowing direct comparison with experimental spectroscopic data both to facilitate interpreting the low-energy region of the spectra and to provide insights into the nature of the bonds themselves. Recent technological development by the authors has realized such computation for the first time. Accurate simulation of the vibrational structure of these compounds has been hampered by their notorious multiconfigurational electronic structure that yields bond distances that do not correlate with bond order. Some measured Cr–Cr vibrational stretching modes, ν(Cr 2 ), have suggested weaker bonding, even for so-called ultrashort Cr–Cr bonds, while others are in line with the bond distance. Here, we optimize geometries and compute ν(Cr 2 ) with CASPT2 for three well-characterized complexes, Cr 2 (O 2 CCH 3 ) 4 (H 2 O) 2 , Cr 2 (mhp) 4 , and Cr 2 (dmp) 4 . We obtain CASPT2 harmonic ν(Cr 2 ) modes in good agreement with experiment at 282 cm –1 for Cr 2 (mhp) 4 and 353 cm –1 for Cr 2 (dmp) 4 , compute 50 Cr and 54 Cr isotope shifts, and demonstrate that the use of the so-called IPEA shift leads to improved Cr–Cr distances. Additionally, normal mode sampling was used to estimate anharmonicity along ν(Cr 2 ), leading to an anharmonic mode of 272 cm –1 for Cr 2 (mhp) 4 and 333 cm –1 for Cr 2 (dmp) 4 .

36 MATERIALS SCIENCE↗

Residual estimation for grid modification in wall-modeled large eddy simulation using unstructured high-order methods

Here, the accuracy and computational cost of a large eddy simulation are highly dependent on the computational grid. Building optimal grids manually from a priori knowledge is not feasible in most practical use cases; instead, solution-adaptive strategies can provide a robust and cost-efficient method to generate a grid with the desired accuracy. We adapt the residual estimation algorithm developed by Toosi and Larsson for Discontinuous Galerkin Spectral Elements Methods (DGSEM) to guide the grid-adaptation process. The core of the method is the computation of the estimated modeling residual using the polynomial basis functions used in DGSEM, and the averaging of the estimated residual over each element. The final method is assessed in multiple channel flow test cases and for the transonic flow over an airfoil, in both cases making use of mortar interfaces between elements with hanging nodes. The method is found to be robust and reliable, and to provide solutions on grids with significantly fewer elements at comparable accuracy compared to when using human-generated grids.

97 MATHEMATICS AND COMPUTING↗

Improving the performance of medical imaging applications using SYCL

As opposed to the Open Computing Language (OpenCL) programming model, in which host and device codes are generally written in different programming languages, SYCL can combine host and device codes for an application in a type-safe way to improve development productivity and performance portability. Hence, this report shows the experimental results of applying the SYCL programming model to medical imaging applications for a study on performance portability and programming productivity between OpenCL and SYCL. Rodinia is a widely used open-source benchmark suite for heterogeneous computing. We choose two medical imaging applications (Heart Wall and Particle Filter) in the benchmark suite, migrate the OpenCL implementations of the applications to the SYCL implementations, and evaluate their performance and productivity on Intel® microprocessors that contain a central processing unit (CPU) and an integrated graphics processing unit (GPU). The maturing SYCL compilers, which are based on a conformant implementation of the SYCL 1.2.1 Khronos specification, have been actively optimized for Intel® computing platforms. The experimental results are promising in terms of the raw performance and productivity. Although the SYCL implementation of the Heart Wall application does not execute successfully on a CPU, the SYCL implementations of the application can achieve comparable or better performance than the OpenCL implementation on Intel® integrated GPUs. For the Particle Filter application, the performance difference between the SYCL and OpenCL implementations is comparable on the GPUs for most cases, but the SYCL implementations are on average 4X faster than the OpenCL implementations on Intel® Xeon® CPUs. For programming productivity, we arguably use lines of code as a way to measure programming productivity in software. The SYCL programs reduce the lines of code of the OpenCL programs by 52% and 38% for the Heart Wall and Particle Filter, respectively. The results indicate that SYCL is a promising programming model for heterogeneous computing with the maturing compilers. We organize the remainder of the report as follows. Section II introduces the SYCL programming model, compares the major differences between an OpenCL program and a SYCL program, and describes the characteristics of the two applications. Section III describes the SYCL programming model in more details, and shows the SYCL implementation of a kernel function in the Particle Filter application as an example. In Section IV, we evaluate the performance of the applications on the CPUs and GPUs. Section V concludes the report.

59 BASIC BIOLOGICAL SCIENCES↗

Cooperative Education

Los Alamos National Laboratory (LANL) is a multidisciplinary national laboratory that conducts research and development in national security, engineering, materials science, computational modeling, and advanced manufacturing. The laboratory develops innovative technologies to address complex scientific and engineering challenges. This project focuses on the development and evaluation of high-performance absorbing structures through computational design, simulation, and engineering analysis. Absorbing structures are used in applications where damage mitigation, structural protection, and material efficiency are critical performance requirements. The increasing demand for lightweight, high-strength, and highly efficient structural systems has created a need for improved design methodologies capable of maximizing absorption while minimizing weight and material usage. The project utilizes advanced engineering software, including 3D CAD software and FEA, to generate and optimize structural concepts. Computational simulations are performed to evaluate structural behavior under loading conditions, while mathematical analyses are conducted using Python-based tools as well as established analytical equations from material and structural mechanics. The project benefits LANL by supporting the development of advanced design methodologies and improving the understanding of material and structural performance. During the internship term, a significant portion of the design development, simulation, and data analysis activities will be completed. Success of the project depends on collaboration among engineering mentors and technical staff members. Work will be conducted at Los Alamos National Laboratory using laboratory computing resources and engineering software.

42 ENGINEERING↗

COEXISTING FLUX ROPE AND DIPPED ARCADE SECTIONS ALONG ONE SOLAR FILAMENT

We compute the three-dimensional magnetic field of an active region in order to study the magnetic configuration of active region filaments. The nonlinear force-free field model is adopted to compute the magnetic field above the photosphere, where the vector magnetic field was observed by THEMIS/MTR on 2005 May 27. We propose a new method to remove the 180{sup 0} ambiguity of the transverse field. Next, we analyze the implications of the preprocessing of the data by minimizing the total force and torque in the observed vector fields. This step provides a consistent bottom boundary condition for the nonlinear force-free field model. Then, using the optimization method to compute the coronal field, we find a magnetic flux rope along the polarity inversion line. The magnetic flux rope aligns well with part of an H{alpha} filament, while the total distribution of the magnetic dips coincides with the whole H{alpha} filament. This implies that the magnetic field structure in one section of the filament is a flux rope, while the other is a sheared arcade. The arcade induced a left-bearing filament in the magnetic field of negative helicity, which is opposite to the chirality of barbs that a flux rope would induce in a magnetic field of the same helicity sign. The field strength in the center of the flux rope is about 700 G, and the twist of the field lines is {approx}1.4 turns.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

N-representability-driven reconstruction of the two-electron reduced-density matrix for a real-time time-dependent electronic structure method

Propagating the equations of motion (EOM) for the one-electron reduced-density matrix (1-RDM) requires knowledge of the corresponding two-electron RDM (2-RDM). We show that the indeterminacy of this expression can be removed through a constrained optimization that resembles the variational optimization of the ground-state 2-RDM subject to a set of known N-representability conditions. Electronic excitation energies can then be obtained by propagating the EOM for the 1-RDM and following the dipole moment after the system interacts with an oscillating external electric field. For simple systems with well-separated excited states whose symmetry differs from that of the ground state, excitation energies obtained from this method are comparable to those obtained from full configuration interaction computations. Although the optimized 2-RDM satisfies necessary N-representability conditions, the procedure cannot guarantee a unique mapping from the 1-RDM to the 2-RDM. This deficiency is evident in the mean-field-quality description of transitions to states of the same symmetry as the ground state, as well as in the inability of the method to describe Rabi oscillations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A spectral-Lagrangian Boltzmann solver for a multi-energy level gas

In this paper a spectral-Lagrangian method is proposed for the full, non-linear Boltzmann equation for a multi-energy level gas typical of a hypersonic re-entry flow. Internal energy levels are treated as separate species and inelastic collisions (leading to internal energy excitation and relaxation) are accounted for. The formulation developed can also be used for the case of a gas mixture made of monatomic gases without internal energy (where only elastic collisions occur). The advantage of the spectral-Lagrangian method lies in the generality of the algorithm in use for the evaluation of the elastic and inelastic collision operators, as well as the conservation of mass, momentum and energy during collisions. The latter is realized through the solution of constrained optimization problems. The computational procedure is based on the Fourier transform of the partial elastic and inelastic collision operators and exploits the fact that these can be written as weighted convolutions in Fourier space with no restriction on the cross-section model. The feasibility of the proposed approach is demonstrated through numerical examples for both space homogeneous and in-homogeneous problems. Computational results are compared with those obtained by means of the DSMC method in order to assess the accuracy of the proposed spectral-Lagrangian method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗