Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel 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

MOCFE-Bone: The 3D-MOC Mini-Application for Exascale Research

MOCFE-Bone is a mini-application that simulates the main procedures in a 3D method of characteristics (MOC) code for numerical solution of the steady state neutron transport equation. 3D MOC was chosen as a specific methodology of interest in exascale computing for nuclear reactor applications because of its heterogeneous geometry capability, high degree of accuracy, and potential for scalability. In developing the mini-app, we focused on emulating the behavior of a production 3D MOC tool without requiring the input overhead associated with such a code. The MOCFE code at Argonne National Laboratory was used to investigate and characterize the behavior of the mini-app; the results of are presented here. Note that the existing MOCFE code resulted from research forays into the potential of 3D-MOC for large scale parallelism and is not a production level tool nor is it optimally implemented. Parts of the mini-app therefore reflect lessons learned and represent how a production-level tool would be written from this point forward rather than simply duplicating MOCFE, which has known scalability issues (i.e. parallelization in energy).

Wolters, E. R.↗

Spacial and objective decompositions for very large SCAPs

This paper considers the single commodity allocation problem (SCAP) for disaster recovery, a fundamental problem faced by all populated areas. SCAPs are complex stochastic optimization problems that combine resource allocation, warehouse routing, and parallel fleet routing. Moreover, these problems must be solved under tight run-time constraints to be practical in real-world disaster situations. This paper revisits the SCAP algorithm proposed in and proposes new storage allocation models that are necessary to enable the algorithm to scale to problem sizes of three orders of magnitude greater (250, 500, 1000 storage locations). The new algorithms are validated on large-scale hurricane disaster scenarios generated by Los Alamos National Laboratory using state-of-the-art disaster simulation tools.

97 MATHEMATICS AND COMPUTING↗

Memory-efficient optimization of Gyrokinetic particle-to-grid interpolation for multicore processors

We present multicore parallelization strategies for the particle-to-grid interpolation step in the Gyrokinetic Toroidal Code (GTC), a 3D particle-in-cell (PIC) application to study turbulent transport in magnetic-confinement fusion devices. Particle-grid interpolation is a known performance bottleneck in several PIC applications. In GTC, this step involves particles depositing charges to a 3D toroidal mesh, and multiple particles may contribute to the charge at a grid point. We design new parallel algorithms for the GTC charge deposition kernel, and analyze their performance on three leading multicore platforms. We implement thirteen different variants for this kernel and identify the best-performing ones given typical PIC parameters such as the grid size, number of particles per cell, and the GTC-specific particle Larmor radius variation. We find that our best strategies can be 2x faster than the reference optimized MPI implementation, and our analysis provides insight into desirable architectural features for high-performance PIC simulation codes.

97 MATHEMATICS AND COMPUTING↗

ParaDyn Implementation in the US Navy's DYSMAS Simulation System: FY08 Progress Report

The goal of this project is to increase the computational efficiency and capacity of the Navy's DYSMAS simulation system for full ship shock response to underwater explosion. Specifically, this project initiates migration to a parallel processing capability for the structural portion of the overall fluid-structure interaction model. The capstone objective for the first phase is to demonstrate operation of the DYSMAS simulation engine with a production model on a Naval Surface Warfare Center (IHD) parallel platform using the ParaDyn code for parallel processing of the structural dynamics. This year saw a successful launch to integrate ParaDyn, the high-parallel structural dynamics code from Lawrence Livermore National Laboratory (LLNL), into the DYSMAS system for simulating the response of ship structures to underwater explosion (UNDEX). The current LLNL version of DYNA3D, representing ten years of general development beyond the source branch used to initiate DYNA-N customization for DYSMAS, was first connected to the GEMINI flow code through DYSMAS Standard Coupler Interface (SCI). This permitted an early 'sanity check' by Naval Surface Warfare Center, Indian Head Division (NSWC-IHD) personnel that equivalent results were generated for their standard UNDEX test problems, thus ensuring the Verification & Validation pedigree they have developed remains intact. The ParaDyn code was then joined to the SCI in a manner requiring no changes to GEMINI. Three NSWC-IHD engineers were twice hosted at LLNL to become familiar with LLNL computer systems, the execution of the prototype software system, and to begin assessment of its accuracy and performance. Scaling data for the flow solver GEMINI was attained up to a one billion cell, 1000 processor run. The NSWC-IHD engineers were granted privileges to continue their evaluations through remote connections to LLNL's Open Computing Facility. Finally, the prototype changes were integrated into the mainline ParaDyn source repository and issued as part of its Version 8.1 beta release. This source was transmitted to NSWC-IHD and in collaboration with LLNL personnel the entire ParaDyn software suite successfully installed and demonstrated on its new SGI Altix machine. The ability of even minor numbers of processors for the structural dynamics to impact overall time-to-solution for DYSMAS has been demonstrated. Assessments of combined parallel efficiencies are beginning to highlight areas for further DYSMAS optimizations.

42 ENGINEERING↗

Vertex Reordering for Real-world Graphs and Applications: An Empirical Evaluation

Vertex reordering is a way to improve locality in graph computations. Given an input (or ``natural'') order, reordering aims to compute an alternate permutation of the vertices that is aimed at maximizing a locality-based objective. Given decades of research on this topic, there are tens of graph reordering schemes, and there are also several linear arrangement ``gap'' measures for treatment as objectives. However, a comprehensive empirical analysis of the efficacy of the ordering schemes against the different gap measures, and against real-world applications is currently lacking. In this study, we present an extensive empirical evaluation of up to 11 ordering schemes, taken from different classes of approaches, on a set of 34 real-world graphs emerging from different application domains. Our study is presented in two parts: a) a thorough comparative evaluation of the different ordering schemes on their effectiveness to optimize different linear arrangement gap measures, relevant to preserving locality; and b) extensive evaluation of the impact of the ordering schemes on two real-world, parallel graph applications, namely, community detection and influence maximization. Our studies show a significant divergence among the ordering schemes (up to $40\times$ between the best and the poor) in their effectiveness to reduce the gap measures; and a wide ranging impact of the ordering schemes on various aspects including application runtime (up to $4\times$), memory and cache use, load balancing, and parallel work and efficiency. The comparative study also help reveal the nuances of a parallel environment (compared to serial) on the ordering schemes and their role in optimizing applications.

Barik, Reet↗

An optimized FM-index library for nucleotide and amino acid search

Abstract Background Pattern matching is a key step in a variety of biological sequence analysis pipelines. The FM-index is a compressed data structure for pattern matching, with search run time that is independent of the length of the database text. Implementation of the FM-index is reasonably complicated, so that increased adoption will be aided by the availability of a fast and flexible FM-index library. Results We present AvxWindowedFMindex (AWFM-index), a lightweight, open-source, thread-parallel FM-index library written in C that is optimized for indexing nucleotide and amino acid sequences. AWFM-index introduces a new approach to storing FM-index data in a strided bit-vector format that enables extremely efficient computation of the FM-index occurrence function via AVX2 bitwise instructions, and combines this with optional on-disk storage of the index’s suffix array and a cache-efficient lookup table for partial k-mer searches. The AWFM-index performs exact match count and locate queries faster than SeqAn3’s FM-index implementation across a range of comparable memory footprints. When optimized for speed, AWFM-index is $$\sim $$ ∼ 2–4x faster than SeqAn3 for nucleotide search, and $$\sim $$ ∼ 2–6x faster for amino acid search; it is also $$\sim $$ ∼ 4x faster with similar memory footprint when storing the suffix array in on-disk SSD storage. Conclusions AWFM-index is easy to incorporate into bioinformatics software, offers run-time performance parameterization, and provides clients with FM-index functionality at both a high-level (count or locate all instances of a query string) and low-level (step-wise control of the FM-index backward-search process). The open-source library is available for download at https://github.com/TravisWheelerLab/AvxWindowFmIndex.

59 BASIC BIOLOGICAL SCIENCES↗

Reductive Analysis with Compiler-Guided Large Language Models for Input-Centric Code Optimizations

Input-centric program optimization aims to optimize code by considering the relations between program inputs and program behaviors. Despite its promise, a long-standing barrier for its adoption is the difficulty of automatically identifying critical features of complex inputs. This paper introduces a novel technique, reductive analysis through compiler-guided Large Language Models (LLMs), to solve the problem through a synergy between compilers and LLMs. It uses a reductive approach to overcome the scalability and other limitations of LLMs in program code analysis. The solution, for the first time, automates the identification of critical input features without heavy instrumentation or profiling, cutting the time needed for input identification by 44× (or 450× for local LLMs), reduced from 9.6 hours to 13 minutes (with remote LLMs) or 77 seconds (with local LLMs) on average, making input characterization possible to be integrated into the workflow of program compilations. Optimizations on those identified input features show similar or even better results than those identified by previous profiling-based methods, leading to optimizations that yield 92.6% accuracy in selecting the appropriate adaptive OpenMP parallelization decisions, and 20-30% performance improvement of serverless computing while reducing resource usage by 50-60%.

Input-Centric Optimization↗

Optimizing checkpoint data placement with guaranteed burst buffer endurance in large-scale hierarchical storage systems

Non-volatile devices, such as SSDs, will be an integral part of the deepening storage hierarchy on large-scale HPC systems. These devices can be on the compute nodes as part of a distributed burst buffer service or they can be external. Wherever they are located in the hierarchy, one critical design issue is the SSD endurance under the write-heavy workloads, such as the checkpoint I/O for scientific applications. For these environments, it is widely assumed that checkpoint operations can occur once every 60 min and for each checkpoint step as much as half of the system memory can be written out. Unfortunately, for large-scale HPC applications, the burst buffer SSDs can be worn out much more quickly given the extensive amount of data written at every checkpoint step. One possible solution is to control the amount of data written by reducing the checkpoint frequency. However, a direct effect caused by reduced checkpoint frequency is the increased vulnerability window of system failures and therefore potentially wasted computation time, especially for large-scale compute jobs.

Computer Science↗

Review of SPECT collimator selection, optimization, and fabrication for clinical and preclinical imaging

In single photon emission computed tomography, the choice of the collimator has a major impact on the sensitivity and resolution of the system. Traditional parallel-hole and fan-beam collimators used in clinical practice, for example, have a relatively poor sensitivity and subcentimeter spatial resolution, while in small-animal imaging, pinhole collimators are used to obtain submillimeter resolution and multiple pinholes are often combined to increase sensitivity. This paper reviews methods for production, sensitivity maximization, and task-based optimization of collimation for both clinical and preclinical imaging applications. New opportunities for improved collimation are now arising primarily because of (i) new collimator-production techniques and (ii) detectors with improved intrinsic spatial resolution that have recently become available. These new technologies are expected to impact the design of collimators in the future. The authors also discuss concepts like septal penetration, high-resolution applications, multiplexing, sampling completeness, and adaptive systems, and the authors conclude with an example of an optimization study for a parallel-hole, fan-beam, cone-beam, and multiple-pinhole collimator for different applications.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

From Physics Model to Results: An Optimizing Framework for Cross-Architecture Code Generation

Starting from a high-level problem description in terms of partial differential equations using abstract tensor notation, the Chemora framework discretizes, optimizes, and generates complete high performance codes for a wide range of compute architectures. Chemora extends the capabilities of Cactus, facilitating the usage of large-scale CPU/GPU systems in an efficient manner for complex applications, without low-level code tuning. Chemora achieves parallelism through MPI and multi-threading, combining OpenMP and CUDA. Optimizations include high-level code transformations, efficient loop traversal strategies, dynamically selected data and instruction cache usage strategies, and JIT compilation of GPU code tailored to the problem characteristics. The discretization is based on higher-order finite differences on multi-block domains. Chemora's capabilities are demonstrated by simulations of black hole collisions. This problem provides an acid test of the framework, as the Einstein equations contain hundreds of variables and thousands of terms.

Blazewicz, Marek↗

Gyrokinetic toroidal simulations on leading multi- and manycore HPC systems

The gyrokinetic Particle-in-Cell (PIC) method is a critical computational tool enabling petascale fusion simulation re-search. In this work, we present novel multi- and manycore-centric optimizations to enhance performance of GTC, a PIC-based production code for studying plasma microtur-bulence in tokamak devices. Our optimizations encompass all six GTC sub-routines and include multi-level particle and grid decompositions designed to improve multi-node parallel scaling, particle binning for improved load balance, GPU acceleration of key subroutines, and memory-centric optimizations to improve single-node scaling and reduce memory utilization. The new hybrid MPI-OpenMP and MPI-OpenMP-CUDA GTC versions achieve up to a 2× speedup over the production Fortran code on four parallel systems - clusters based on the AMD Magny-Cours, Intel Nehalem-EP, IBM BlueGene/P, and NVIDIA Fermi architectures. Finally, strong scaling experiments provide insight into parallel scalability, memory utilization, and programmability trade-offs for large-scale gyrokinetic PIC simulations, while attaining a 1.6× speedup on 49,152 XE6 cores.

97 MATHEMATICS AND COMPUTING↗

A novel electron gun for inline MRI-linac configurations

Purpose: This work introduces a new electron gun geometry capable of robust functioning in the presence of a high strength external magnetic field for axisymmetric magnetic resonance imaging (MRI)-linac configurations. This allows an inline MRI-linac to operate without the need to isolate the linear accelerator (linac) using a magnetic shield. This MRI-linac integration approach not only leaves the magnet homogeneity unchanged but also provides the linac flexibility to move along the magnet axis of symmetry if the source to target distance needs to be adjusted. Methods: Simple electron gun geometry modifications of a Varian 600C electron gun are considered and solved in the presence of an external magnetic field in order to determine a set of design principles for the new geometry. Based on these results, a new gun geometry is proposed and optimized in the fringe field of a 0.5 T open bore MRI magnet (GE Signa SP). A computer model for the 6 MeV Varian 600C linac is used to determine the capture efficiency of the new electron gun-linac system in the presence of the fringe field of the same MRI scanner. The behavior of the new electron gun plus the linac system is also studied in the fringe fields of two other magnets, a 1.0 T prototype open bore magnet and a 1.5 T GE Conquest scanner. Results: Simple geometrical modifications of the original electron gun geometry do not provide feasible solutions. However, these tests show that a smaller transverse cathode diameter with a flat surface and a slightly larger anode diameter could alleviate the current loss due to beam interactions with the anode in the presence of magnetic fields. Based on these findings, an initial geometry resembling a parallel plate capacitor with a hole in the anode is proposed. The optimization procedure finds a cathode-anode distance of 5 mm, a focusing electrode angle of 5°, and an anode drift tube length of 17.1 mm. Also, the linac can be displaced with ±15 cm along the axis of the 0.5 T magnet without capture efficiency reduction below the experimental value in zero field. In this range of linac displacements, the electron beam generated by the new gun geometry is more effectively injected into the linac in the presence of an external magnetic field, resulting in approximately 20% increase of the target current compared to the original gun geometry behavior at zero field. The new gun geometry can generate and accelerate electron beams in external magnetic fields without current loss for fields higher than 0.11 T. The new electron-gun geometry is robust enough to function in the fringe fields of the other two magnets with a target current loss of no more than 16% with respect to the current obtained with no external magnetic fields. Conclusions: In this work, a specially designed electron gun was presented which can operate in the presence of axisymmetric strong magnetic fringe fields of MRI magnets. Computer simulations show that the electron gun can produce high quality beams which can be injected into a straight through linac such as Varian 600C and accelerated with more efficiency in the presence of the external magnetic fields. Also, the new configuration allows linac displacements along the magnet axis in a range equal to the diameter of the imaging spherical volume of the magnet under consideration. The new electron gun-linac system can function in the fringe field of a MRI magnet if the field strength at the cathode position is higher than 0.11 T. The capture efficiency of the linac depends on the magnetic field strength and the field gradient. The higher the gradient the better the capture efficiency. The capture efficiency does not degrade more than 16%.

43 PARTICLE ACCELERATORS↗

Parallelizing Single Source Shortest Path with OpenSHMEM

Single Source Shortest Path (SSSP) is one of the widely occurring graph problems where the paths are discovered from an origin vertex to all other vertices in the graph. In this paper, we discuss our experience parallelizing SSSP using OpenSHMEM. We start with the serial Dijkstra and Bellman-Ford algorithms, parallelize these algorithms, and adapt them to the Partitioned Global Address Space (PGAS) programming model. We implement the parallel algorithms using OpenSHMEM and introduce a series of optimizations to achieve higher scaling and performance characteristics. The implementation is evaluated on Titan with various graphs including synthetic Recursive Matrix (R-MAT) and small-world network graphs as well as real-world graphs from Facebook, Twitter, LiveJournal, and the road maps of California and Texas.

97 MATHEMATICS AND COMPUTING↗

Mechanisms of nickel sorption by a bacteriogenic birnessite

A synergistic experimental-computational approach was used to study the molecular-scale mechanisms of Ni sorption at varying loadings and at pH 6-8 on the biogenic hexagonal birnessite produced by Pseudomonas putida GB-1. We found that Ni is scavenged effectively by bacterial biomass-birnessite assemblages. At surface excess values below 0.18 mol Ni kg{sup -1} sorbent (0.13 mol Ni mol{sup -1} Mn), the biomass component of the sorbent did not interfere with Ni sorption on mineral sites. Extended X-ray absorption fine structure (EXAFS) spectra showed two dominant coordination environments: Ni bound as a triple-corner-sharing (Ni-TCS) complex at vacancy sites and Ni incorporated (Ni-inc) into the MnO{sub 2} sheet, with the latter form of Ni favored at high sorptive concentrations and decreased proton activity. In parallel to our spectral analysis, first-principles geometry optimizations based on density functional theory (DFT) were performed to investigate the structure of Ni surface complexes at vacancy sites. Excellent agreement was achieved between EXAFS- and DFT-derived structural parameters for Ni-TCS and Ni-inc. Reaction-path calculations revealed a pH-dependent energy barrier associated with the transition from Ni-TCS to Ni-inc. Our results are consistent with the rate-limited incorporation of Ni at vacancy sites in our sorption samples, but near-equilibrium state of Ni in birnessite phases found in nodule samples. This study thus provides direct and quantitative evidence of the factors governing the occurrence of Ni adsorption versus Ni incorporation in biogenic hexagonal birnessite, a key mineral in the terrestrial manganese cycle.

54 ENVIRONMENTAL SCIENCES↗

Distributed memory, GPU accelerated Fock construction for hybrid, Gaussian basis density functional theory

With the growing reliance of modern supercomputers on accelerator-based architecture such a graphics processing units (GPUs), the development and optimization of electronic structure methods to exploit these massively parallel resources has become a recent priority. While significant strides have been made in the development GPU accelerated, distributed memory algorithms for many modern electronic structure methods, the primary focus of GPU development for Gaussian basis atomic orbital methods has been for shared memory systems with only a handful of examples pursing massive parallelism. Here in this work, we present a set of distributed memory algorithms for the evaluation of the Coulomb and exact exchange matrices for hybrid Kohn–Sham DFT with Gaussian basis sets via direct density-fitted (DF-J-Engine) and seminumerical (sn-K) methods, respectively. The absolute performance and strong scalability of the developed methods are demonstrated on systems ranging from a few hundred to over one thousand atoms using up to 128 NVIDIA A100 GPUs on the Perlmutter supercomputer.

97 MATHEMATICS AND COMPUTING↗

Automatic Detection and Evaluation of Solar Cell Micro-Cracks in Electroluminescence Images Using Matched Filters

A method for detecting micro-cracks in solar cells using two dimensional matched filters was developed, derived from the electroluminescence intensity profile of typical micro-cracks. We describe the image processing steps to obtain a binary map with the location of the micro-cracks. Finally, we show how to automatically estimate the total length of each micro-crack from these maps, and propose a method to identify severe types of micro-cracks, such as parallel, dendritic, and cracks with multiple orientations. With an optimized threshold parameter, the technique detects over 90 % of cracks larger than 3 cm in length. The method shows great potential for quantifying micro-crack damage after manufacturing or module transportation for the determination of a module quality criterion for cell cracking in photovoltaic modules.

14 SOLAR ENERGY↗

Progress and R/D challenges for FCC-ee SRF

The FCC-ee machines present a huge challenge for the RF systems, which need to be adapted to very diverse beam conditions going from moderate energy and high current for the Z machine to high energy and low beam current for the ttbar. This inverse scaling results naturally from a fixed budget for the synchrotron radiation, which the SRF cavities need to compensate. A global solution was elaborated for the FCC Conceptual Design Report (Abada in Eur Phys J Spec Top 228):261–623, 2019), and is referred here as the baseline. Recently, further studies have led to a new optimized baseline, still based on traditional elliptical cavities. In parallel, a novel concept, named the Slotted Waveguide ELLiptical (SWELL), was proposed with the potential of greatly simplified logistics and reduced costs. Under several aspects, all these changes call for enhanced performance of the RF systems. A vigorous R&D program has therefore continued since the publication of the CDR, with the aim of pushing the performance and demonstrating the feasibility of a more advanced baseline and, more recently, of the SWELL option. The progress and challenges of this ambitious program were presented in the dedicated SRF sessions at FCC week 2022 and are summarized in this paper.

47 OTHER INSTRUMENTATION↗

Galactos: computing the anisotropic 3-point correlation function for 2 billion galaxies

The nature of dark energy and the complete theory of gravity are two central questions currently facing cosmology. A vital tool for addressing them is the 3-point correlation function (3PCF), which probes deviations from a spatially random distribution of galaxies. However, the 3PCF's formidable computational expense has prevented its application to astronomical surveys comprising millions to billions of galaxies. We present Galactos, a high-performance implementation of a novel, O( N 2 ) algorithm that uses a load-balanced k-d tree and spherical harmonic expansions to compute the anisotropic 3PCF. Our implementation is optimized for the Intel Xeon Phi architecture, exploiting SIMD parallelism, instruction and thread concurrency, and significant L1 and L2 cache reuse, reaching 39% of peak performance on a single node. Galactos scales to the full Cori system, achieving 9.8 PF (peak) and 5.06 PF (sustained) across 9636 nodes, making the 3PCF easily computable for all galaxies in the observable universe.

79 ASTRONOMY AND ASTROPHYSICS↗