Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel algorithm”

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 1,279 records · Page 71
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

GraphReduce: Processing Large-Scale Graphs on Accelerator-Based Systems

Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device’s internal memory capacity. GraphReduce adopts a combination of edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and device.

Architecture optimization, reuse, performance, ene↗
DOE OSTI2024

Data-flow parallelism for high-energy and nuclear physics frameworks

The processing tasks of an event-processing workflow in high-energy and nuclear physics (HENP) can typically be represented as a directed acyclic graph formed according to the data flow—i.e. the data dependencies among algorithms executed as part of the workflow. With this representation, an HENP framework can optimally execute a workflow, exploiting the parallelism inherent among independent tasks. Despite such a natural description of a workflow, most HENP frameworks do not make use of technologies that provide concurrent execution of graph-based tasking structures. In this talk, we describe Fermilab efforts to adopt a graph-based technology (specifically Intel’s oneTBB flow graph) for meeting the framework needs of its experiments, notably DUNE. Building on the Meld project as presented at CHEP2023, we demonstrate that all common processing idioms supported by current frameworks can naturally be supported by oneTBB’s data-flow technology, optimally leveraging the concurrent capabilities of the machine. In addition, we discuss collaborative efforts between Fermilab and the Intel oneTBB development team, who is considering improvements to the flow-graph technology to better support HENP use cases.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2015

GraphReduce: Large-Scale Graph Analytics on Accelerator-Based HPC Systems

Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device’s internal memory capacity. GraphReduce adopts a combination of both edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and the device.

Architecture optimization, reuse, performance, ene↗
DOE OSTI2011

Progress report on Nuclear Density project with Lawrence Livermore National Lab Year 2010

The main goal for year 2010 was to improve parallelization of the configuration interaction code BIGSTICK, co-written by W. Erich Ormand (LLNL) and Calvin W. Johnson (SDSU), with the parallelization carried out primarily by Plamen Krastev, a postdoc at SDSU and funded in part by this grant. The central computational algorithm is the Lanczos algorithm, which consists of a matrix-vector multiplication (matvec), followed by a Gram-Schmidt reorthogonalization.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2014

SpArcFiRe: Scalable automated detection of spiral galaxy arm segments

Given an approximately centered image of a spiral galaxy, we describe an entirely automated method that finds, centers, and sizes the galaxy (possibly masking nearby stars and other objects if necessary in order to isolate the galaxy itself) and then automatically extracts structural information about the spiral arms. For each arm segment found, we list the pixels in that segment, allowing image analysis on a per-arm-segment basis. We also perform a least-squares fit of a logarithmic spiral arc to the pixels in that segment, giving per-arc parameters, such as the pitch angle, arm segment length, location, etc. The algorithm takes about one minute per galaxies, and can easily be scaled using parallelism. We have run it on all ∼644,000 Sloan objects that are larger than 40 pixels across and classified as 'galaxies'. We find a very good correlation between our quantitative description of a spiral structure and the qualitative description provided by Galaxy Zoo humans. Our objective, quantitative measures of structure demonstrate the difficulty in defining exactly what constitutes a spiral 'arm', leading us to prefer the term 'arm segment'. We find that pitch angle often varies significantly segment-to-segment in a single spiral galaxy, making it difficult to define the pitch angle for a single galaxy. We demonstrate how our new database of arm segments can be queried to find galaxies satisfying specific quantitative visual criteria. For example, even though our code does not explicitly find rings, a good surrogate is to look for galaxies having one long, low-pitch-angle arm—which is how our code views ring galaxies. SpArcFiRe is available at http://sparcfire.ics.uci.edu.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Immersed Boundary Method Halo Exchange in a Hemodynamics Application

In recent years, highly parallelized simulations of blood flow resolving individual blood cells have been demonstrated. Simulating such dense suspensions of deformable particles in flow often involves a partitioned fluid-structure interaction (FSI) algorithm, with separate solvers for Eulerian fluid and Lagrangian cell grids, plus a solver - e.g., immersed boundary method - for their interaction. Managing data motion in parallel FSI implementations is increasingly important, particularly for inhomogeneous systems like vascular geometries. In this study, we evaluate the influence of Eulerian and Lagrangian halo exchanges on efficiency and scalability of a partitioned FSI algorithm for blood flow. We describe an MPI+OpenMP implementation of the immersed boundary method coupled with lattice Boltzmann and finite element methods. We consider how communication and recomputation costs influence the optimization of halo exchanges with respect to three factors: immersed boundary interaction distance, cell suspension density, and relative fluid/cell solver costs.

Gounley, John↗
DOE OSTI2013

Scalable load balancing for massively parallel distributed Monte Carlo particle transport

In order to run computer simulations efficiently on massively parallel computers with hundreds of thousands or millions of processors, care must be taken that the calculation is load balanced across the processors. Examining the workload of every processor leads to an unscalable algorithm, with run time at least as large as O(N), where N is the number of processors. We present a scalable load balancing algorithm, with run time 0(log(N)), that involves iterated processor-pair-wise balancing steps, ultimately leading to a globally balanced workload. We demonstrate scalability of the algorithm up to 2 million processors on the Sequoia supercomputer at Lawrence Livermore National Laboratory. (authors)

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

A Flexible Forwarding Scheme to Improve Latency-Bound Irregular P2P Communication in MPI

We propose an algorithm to efficiently perform latency-bound communication scenarios that consist of many small messages. In these parallel scenarios, processes typically pass around a lot of small-sized messages of a few KBs of size. Performing communication operations with P2P MPI routines or collective MPI routines (including neighborhood collectives) in such scenarios may not always yield the optimal results and may not resolve the latency bottleneck. To this end, we develop a regular structure called virtual process topology (VPT) on which the messages can be communicated in a structured and controlled manner. Using parameters of this topology, one can tune the rate of aggression in tackling the latency costs. We demonstrate that our communication algorithm is preferable to MPI P2P and collective routines for latency-bound communication and it can easily be adapted only by replacing calls to MPI routines in a parallel application. We show how to adapt existing topology-aware mapping heuristics to address the volume overhead due to communicating messages on the VPT. Moreover, we propose a novel swap-based mapping heuristic to address this overhead by optimizing the maximum volume handled by a process. Experiments on synthetic communication graphs as well as real-world applications such as parallel Canonical Polyadic sparse tensor decomposition and parallel sparse matrix-dense matrix multiplication show that our approach is a powerful way of overcoming the bottlenecks posed by sparse and latency-bound irregular communication.

communication algorithm↗
DOE OSTI2017

Information criteria for quantifying loss of reversibility in parallelized KMC

Parallel Kinetic Monte Carlo (KMC) is a potent tool to simulate stochastic particle systems efficiently. However, despite literature on quantifying domain decomposition errors of the particle system for this class of algorithms in the short and in the long time regime, no study yet explores and quantifies the loss of time-reversibility in Parallel KMC. Inspired by concepts from non-equilibrium statistical mechanics, we propose the entropy production per unit time, or entropy production rate, given in terms of an observable and a corresponding estimator, as a metric that quantifies the loss of reversibility. Typically, this is a quantity that cannot be computed explicitly for Parallel KMC, which is why we develop a posteriori estimators that have good scaling properties with respect to the size of the system. Through these estimators, we can connect the different parameters of the scheme, such as the communication time step of the parallelization, the choice of the domain decomposition, and the computational schedule, with its performance in controlling the loss of reversibility. From this point of view, the entropy production rate can be seen both as an information criterion to compare the reversibility of different parallel schemes and as a tool to diagnose reversibility issues with a particular scheme. As a demonstration, we use Sandia Lab's SPPARKS software to compare different parallelization schemes and different domain (lattice) decompositions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

An algorithm for continuum modeling of rocks with multiple embedded nonlinearly-compliant joints [Continuum modeling of elasto-plastic media with multiple embedded nonlinearly-compliant joints]

Here, we present a numerical method for modeling the mechanical effects of nonlinearly-compliant joints in elasto-plastic media. The method uses a series of strain-rate and stress update algorithms to determine joint closure, slip, and solid stress within computational cells containing multiple “embedded” joints. This work facilitates efficient modeling of nonlinear wave propagation in large spatial domains containing a large number of joints that affect bulk mechanical properties. We implement the method within the massively parallel Lagrangian code GEODYN-L and provide verification and examples. We highlight the ability of our algorithms to capture joint interactions and multiple weakness planes within individual computational cells, as well as its computational efficiency. We also discuss the motivation for developing the proposed technique: to simulate large-scale wave propagation during the Source Physics Experiments (SPE), a series of underground explosions conducted at the Nevada National Security Site (NNSS).

42 ENGINEERING↗
DOE OSTI2010

The OVIS analysis architecture.

This report summarizes the current statistical analysis capability of OVIS and how it works in conjunction with the OVIS data readers and interpolators. It also documents how to extend these capabilities. OVIS is a tool for parallel statistical analysis of sensor data to improve system reliability. Parallelism is achieved using a distributed data model: many sensors on similar components (metaphorically sheep) insert measurements into a series of databases on computers reserved for analyzing the measurements (metaphorically shepherds). Each shepherd node then processes the sheep data stored locally and the results are aggregated across all shepherds. OVIS uses the Visualization Tool Kit (VTK) statistics algorithm class hierarchy to perform analysis of each process's data but avoids VTK's model aggregation stage which uses the Message Passing Interface (MPI); this is because if a single process in an MPI job fails, the entire job will fail. Instead, OVIS uses asynchronous database replication to aggregate statistical models. OVIS has several additional features beyond those present in VTK that, first, accommodate its particular data format and, second, improve the memory and speed of the statistical analyses. First, because many statistical algorithms are multivariate in nature and sensor data is typically univariate, interpolation of data is required to provide simultaneous observations of metrics. Note that in this report, we will refer to a single value obtained from a sensor as a measurement while a collection of multiple sensor values simultaneously present in the system is an observation. A base class for interpolation is provided that abstracts the operation of converting multiple sensor measurements into simultaneous observations. A concrete implementation is provided that performs piecewise constant temporal interpolation of multiple metrics across a single component. Secondly, because calculations may summarize data too large to fit in memory OVIS analyses batches of observations at a time and aggregates these intermediate intra-process models as it goes before storing the final model for inter-process aggregation via database replication. This reduces the memory footprint of the analysis, interpolation, and the database client and server query processing. This also interleaves processing with the disk I/O required to fetch data from the database - also improving speed. This report documents how OVIS performs analyses and how to create additional analysis components that fetch measurements from the database, perform interpolation, or perform operations on streamed observations (such as model updates or assessments). The rest of this section outlines the OVIS analysis algorithm and is followed by sections specific to each subtask. Note that we are limiting our discussion for now to the creation of a model from a set of measurements, and not including the assessment of observations using a model. The same framework can be used for assessment but that use case is not detailed in this report.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Understanding performance variability in standard and pipelined parallel Krylov solvers

In this work, we collect data from runs of Krylov subspace methods and pipelined Krylov algorithms in an effort to understand and model the impact of machine noise and other sources of variability on performance. We find large variability of Krylov iterations between compute nodes for standard methods that is reduced in pipelined algorithms, directly supporting conjecture, as well as large variation between statistical distributions of runtimes across iterations. Based on these results, we improve upon a previously introduced nondeterministic performance model by allowing iterations to fluctuate over time. We present our data from runs of various Krylov algorithms across multiple platforms as well as our updated non-stationary model that provides good agreement with observations. We also suggest how it can be used as a predictive tool.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Data-flow parallelism for high-energy and nuclear physics computing frameworks

The processing tasks of a scientific workflow in high-energy and nuclear physics (HENP) can typically be represented as a directed acyclic graph formed according to the data flow—i.e. the data dependencies among algorithms executed as part of the workflow. With this representation, an HENP computing framework can optimally execute a workflow, exploiting the parallelism inherent among independent tasks. Despite such a natural description of a workflow, most HENP frameworks do not make use of technologies that provide concurrent execution of graph-based tasking structures. In this session, we describe Fermilab efforts to adopt a graph-based technology (specifically Intel’s oneTBB flow graph) for meeting the framework needs of its experiments, notably DUNE. After introducing the physics DUNE intends to explore, we will show that all common processing idioms supported by current HENP frameworks can naturally be supported by oneTBB’s data-flow technology, optimally leveraging the concurrent capabilities of the machine. In addition, we discuss collaborative efforts between Fermilab and the Intel oneTBB development team, who is considering improvements to the flow-graph technology to better support HENP use cases.

43 PARTICLE ACCELERATORS↗
DOE OSTI2011

Intensity-Modulated Radiotherapy for Locally Advanced Non-Small-Cell Lung Cancer: A Dose-Escalation Planning Study

Purpose: To evaluate the potential for dose escalation with intensity-modulated radiotherapy (IMRT) in positron emission tomography-based radiotherapy planning for locally advanced non-small-cell lung cancer (LA-NSCLC). Methods and Materials: For 35 LA-NSCLC patients, three-dimensional conformal radiotherapy and IMRT plans were made to a prescription dose (PD) of 66 Gy in 2-Gy fractions. Dose escalation was performed toward the maximal PD using secondary endpoint constraints for the lung, spinal cord, and heart, with de-escalation according to defined esophageal tolerance. Dose calculation was performed using the Eclipse pencil beam algorithm, and all plans were recalculated using a collapsed cone algorithm. The normal tissue complication probabilities were calculated for the lung (Grade 2 pneumonitis) and esophagus (acute toxicity, grade 2 or greater, and late toxicity). Results: IMRT resulted in statistically significant decreases in the mean lung (p <.0001) and maximal spinal cord (p = .002 and 0005) doses, allowing an average increase in the PD of 8.6-14.2 Gy (p {<=}.0001). This advantage was lost after de-escalation within the defined esophageal dose limits. The lung normal tissue complication probabilities were significantly lower for IMRT (p <.0001), even after dose escalation. For esophageal toxicity, IMRT significantly decreased the acute NTCP values at the low dose levels (p = .0009 and p <.0001). After maximal dose escalation, late esophageal tolerance became critical (p <.0001), especially when using IMRT, owing to the parallel increases in the esophageal dose and PD. Conclusion: In LA-NSCLC, IMRT offers the potential to significantly escalate the PD, dependent on the lung and spinal cord tolerance. However, parallel increases in the esophageal dose abolished the advantage, even when using collapsed cone algorithms. This is important to consider in the context of concomitant chemoradiotherapy schedules using IMRT.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2017

Late-time growth rate, mixing, and anisotropy in the multimode narrowband Richtmyer–Meshkov instability: The θ-group collaboration

Turbulent Richtmyer–Meshkov instability (RMI) is investigated through a series of high resolution three-dimensional simulations of two initial conditions with eight independent codes. The simulations are initialised with a narrowband perturbation such that instability growth is due to non-linear coupling/backscatter from the energetic modes, thus generating the lowest expected growth rate from a pure RMI. By independently assessing the results from each algorithm and computing ensemble averages of multiple algorithms, the results allow a quantification of key flow properties as well as the uncertainty due to differing numerical approaches. A new analytical model predicting the initial layer growth for a multimode narrowband perturbation is presented, along with two models for the linear and non-linear regimes combined. Overall, the growth rate exponent is determined as θ=0.292±0.009, in good agreement with prior studies; however, the exponent is decaying slowly in time. Also, θ is shown to be relatively insensitive to the choice of mixing layer width measurements. The asymptotic integral molecular mixing measures Θ=0.792±0.014, Ξ=0.800±0.014, and Ψ=0.782±0.013 are lower than some experimental measurements but within the range of prior numerical studies. The flow field is shown to be persistently anisotropic for all algorithms, at the latest time having between 49% and 66% higher kinetic energy in the shock parallel direction compared to perpendicular and does not show any return to isotropy. The plane averaged volume fraction profiles at different time instants collapse reasonably well when scaled by the integral width, implying that the layer can be described by a single length scale and thus a single θ. In conclusion, quantitative data given for both ensemble averages and individual algorithms provide useful benchmark results for future research.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Evaluation of Graph Analytics Frameworks Using the GAP Benchmark Suite

The analysis of connected data is an increasingly important application in high-performance computing. Such analyses can reveal fraudulent patterns in financial transactions, optimize telecommunications networks, predict information flow in social networks, etc. However, the landscape of graph analytics is highly diverse. Graph algorithms stress processor architectures differently, and no one graph can represent all topologies. Consequently, no single approach or framework is expected to be optimal for all graph analytics problems. To help make sense of this diverse landscape, we evaluated four approaches to graph analytics: GraphBLAS, Galois, BGL17, GraphIt; and compare them against hand-tuned implementations that take advantage of hardware features on our test platform. Graph- BLAS formulates graph analytics as sparse linear algebra. Galois provides syntactic constructs for data parallelism over irregular data structures. BGL17 is a generic C++ template library for implementing graph algorithms. GraphIt provides a domain- specific language to describe and optimize graph algorithms. We use the GAP Benchmark Suite to establish baseline performance and guide the side-by-side evaluation of each framework. GAP consists of 30 tests: six graph analytics algorithms (breadth- first search, single-source shortest path, PageRank, betweenness centrality, connected components, and triangle counting) run on five graphs, each with different topological characteristics (e.g., high diameter, skewed degree distribution, high average degree). High-performance reference implementations are included for each benchmark algorithm. Because a graph can be loaded into memory a number of ways (e.g., flat file on disk, compressed sparse format, data frames, retrieved from SQL or NoSQL databases), our evaluation focused on computational performance rather than I/O. Our results show the relative strengths of each framework.

Graph algorithms, Benchmarking, shared-memory prog↗
DOE OSTI2021

Twelve Ways to Fool the Masses When Giving Parallel-in-Time Results

Getting good speedup—let alone high parallel efficiency—for parallel-in-time (PinT) integration examples can be frustratingly difficult. The high complexity and large number of parameters in PinT methods can easily (and unintentionally) lead to numerical experiments that overestimate the algorithm’s performance. In the tradition of Bailey’s article “Twelve ways to fool the masses when giving performance results on parallel computers”, we discuss and demonstrate pitfalls to avoid when evaluating the performance of PinT methods. Despite being written in a light-hearted tone, this paper is intended to raise awareness that there are many ways to unintentionally fool yourself and others and that by avoiding these fallacies more meaningful PinT performance results can be obtained.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

A Pervasive Parallel Processing Framework for Data Visualization and Analysis at Extreme Scale

The evolution of the computing world from teraflop to petaflop has been relatively effortless, with several of the existing programming models scaling effectively to the petascale. The migration to exascale, however, poses considerable challenges. All industry trends infer that the exascale machine will be built using processors containing hundreds to thousands of cores per chip. It can be inferred that efficient concurrency on exascale machines requires a massive amount of concurrent threads, each performing many operations on a localized piece of data. Currently, visualization libraries and applications are based off what is known as the visualization pipeline. In the pipeline model, algorithms are encapsulated as filters with inputs and outputs. These filters are connected by setting the output of one component to the input of another. Parallelism in the visualization pipeline is achieved by replicating the pipeline for each processing thread. This works well for today’s distributed memory parallel computers but cannot be sustained when operating on processors with thousands of cores. Our project investigates a new visualization framework designed to exhibit the pervasive parallelism necessary for extreme scale machines. Our framework achieves this by defining algorithms in terms of worklets, which are localized stateless operations. Worklets are atomic operations that execute when invoked unlike filters, which execute when a pipeline request occurs. The worklet design allows execution on a massive amount of lightweight threads with minimal overhead. Only with such fine-grained parallelism can we hope to fill the billions of threads we expect will be necessary for efficient computation on an exascale machine.

97 MATHEMATICS AND COMPUTING↗