Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “multithreading”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

AN MHD AVALANCHE IN A MULTI-THREADED CORONAL LOOP

For the first time, we demonstrate how an MHD avalanche might occur in a multithreaded coronal loop. Considering 23 non-potential magnetic threads within a loop, we use 3D MHD simulations to show that only one thread needs to be unstable in order to start an avalanche even when the others are below marginal stability. This has significant implications for coronal heating in that it provides for energy dissipation with a trigger mechanism. The instability of the unstable thread follows the evolution determined in many earlier investigations. However, once one stable thread is disrupted, it coalesces with a neighboring thread and this process disrupts other nearby threads. Coalescence with these disrupted threads then occurs leading to the disruption of yet more threads as the avalanche develops. Magnetic energy is released in discrete bursts as the surrounding stable threads are disrupted. The volume integrated heating, as a function of time, shows short spikes suggesting that the temporal form of the heating is more like that of nanoflares than of constant heating.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

HOOKED FLARE RIBBONS AND FLUX-ROPE-RELATED QSL FOOTPRINTS

We studied the magnetic topology of active region 12158 on 2014 September 10 and compared it with the observations before and early in the flare that begins at 17:21 UT (SOL2014-09-10T17:45:00). Our results show that the sigmoidal structure and flare ribbons of this active region observed by the Solar Dynamics Observatory /Atmospheric Imaging Assembly can be well reproduced from a Grad–Rubin nonlinear force-free field extrapolation method. Various inverse-S- and inverse-J-shaped magnetic field lines, which surround a coronal flux rope, coincide with the sigmoid as observed in different extreme-ultraviolet wavelengths, including its multithreaded curved ends. Also, the observed distribution of surface currents in the magnetic polarity where it was not prescribed is well reproduced. This validates our numerical implementation and setup of the Grad–Rubin method. The modeled double inverse-J-shaped quasi-separatrix layer (QSL) footprints match the observed flare ribbons during the rising phase of the flare, including their hooked parts. The spiral-like shape of the latter may be related to a complex pre-eruptive flux rope with more than one turn of twist, as obtained in the model. These ribbon-associated flux-rope QSL footprints are consistent with the new standard flare model in 3D, with the presence of a hyperbolic flux tube located below an inverse-teardrop-shaped coronal QSL. This is a new step forward forecasting the locations of reconnection and ribbons in solar flares and the geometrical properties of eruptive flux ropes.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

DATA-DRIVEN RADIATIVE HYDRODYNAMIC MODELING OF THE 2014 MARCH 29 X1.0 SOLAR FLARE

Spectroscopic observations of solar flares provide critical diagnostics of the physical conditions in the flaring atmosphere. Some key features in observed spectra have not yet been accounted for in existing flare models. Here we report a data-driven simulation of the well-observed X1.0 flare on 2014 March 29 that can reconcile some well-known spectral discrepancies. We analyzed spectra of the flaring region from the Interface Region Imaging Spectrograph ( IRIS ) in Mg ii h and k, the Interferometric BIdimensional Spectropolarimeter at the Dunn Solar Telescope (DST/IBIS) in H α 6563 Å and Ca ii 8542 Å, and the Reuven Ramaty High Energy Solar Spectroscope Imager ( RHESSI ) in hard X-rays. We constructed a multithreaded flare loop model and used the electron flux inferred from RHESSI data as the input to the radiative hydrodynamic code RADYN to simulate the atmospheric response. We then synthesized various chromospheric emission lines and compared them with the IRIS and IBIS observations. In general, the synthetic intensities agree with the observed ones, especially near the northern footpoint of the flare. The simulated Mg ii line profile has narrower wings than the observed one. This discrepancy can be reduced by using a higher microturbulent velocity (27 km s{sup −1}) in a narrow chromospheric layer. In addition, we found that an increase of electron density in the upper chromosphere within a narrow height range of ≈800 km below the transition region can turn the simulated Mg ii line core into emission and thus reproduce the single peaked profile, which is a common feature in all IRIS flares.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

Transition region and chromospheric signatures of impulsive heating events. II. Modeling

Results from the Solar Maximum Mission showed a close connection between the hard X-ray (HXR) and transition region (TR) emission in solar flares. Analogously, the modern combination of RHESSI and IRIS data can inform the details of heating processes in ways that were never before possible. We study a small event that was observed with RHESSI, IRIS, SDO, and Hinode, allowing us to strongly constrain the heating and hydrodynamical properties of the flare, with detailed observations presented in a previous paper. Long duration redshifts of TR lines observed in this event, as well as many other events, are fundamentally incompatible with chromospheric condensation on a single loop. We combine RHESSI and IRIS data to measure the energy partition among the many magnetic strands that comprise the flare. Using that observationally determined energy partition, we show that a proper multithreaded model can reproduce these redshifts in magnitude, duration, and line intensity, while simultaneously being well constrained by the observed density, temperature, and emission measure. We comment on the implications for both RHESSI and IRIS observations of flares in general, namely that: (1) a single loop model is inconsistent with long duration redshifts, among other observables; (2) the average time between energization of strands is less than 10 s, which implies that for a HXR burst lasting 10 minutes, there were at least 60 strands within a single IRIS pixel located on the flare ribbon; (3) the majority of these strands were explosively heated with an energy distribution well described by a power law of slope ≈−1.6; (4) the multi-stranded model reproduces the observed line profiles, peak temperatures, differential emission measure distributions, and densities.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

Enabling Graph Appliance for Genome Assembly

In recent years, there has been a huge growth in the amount of genomic data available as reads generated from various genome sequencers. The number of reads generated can be huge, ranging from hundreds to billions of nucleotide, each varying in size. Assembling such large amounts of data is one of the challenging computational problems for both biomedical and data scientists. Most of the genome assemblers developed have used de Bruijn graph techniques. A de Bruijn graph represents a collection of read sequences by billions of vertices and edges, which require large amounts of memory and computational power to store and process. This is the major drawback to de Bruijn graph assembly. Massively parallel, multi-threaded, shared memory systems can be leveraged to overcome some of these issues. The objective of our research is to investigate the feasibility and scalability issues of de Bruijn graph assembly on Cray s Urika-GD system; Urika-GD is a high performance graph appliance with a large shared memory and massively multithreaded custom processor designed for executing SPARQL queries over large-scale RDF data sets. However, to the best of our knowledge, there is no research on representing a de Bruijn graph as an RDF graph or finding Eulerian paths in RDF graphs using SPARQL for potential genome discovery. In this paper, we address the issues involved in representing a de Bruin graphs as RDF graphs and propose an iterative querying approach for finding Eulerian paths in large RDF graphs. We evaluate the performance of our implementation on real world ebola genome datasets and illustrate how genome assembly can be accomplished with Urika-GD using iterative SPARQL queries.

Big data↗
DOE OSTI2014

Framework for Network Co-Simulation

The Framework for Network Co-Simulation (FNCS) uses a federated approach to integrate simulations which may have differing time scales. Special consideration is given to integration with a communication network simulation such that inter-simulation messages may be optionally routed through and delayed by such a simulation. In addition, FNCS uses novel time synchronization algorithms to accelerate co-simulation including the application of speculative multithreading. FNCS accomplishes all of these improvements with minimal end user intervention. Simulations can be integrated using FNCS while maintaining their original model input files simply by linking with the FNCS library and making appropriate calls into the FNCS API.

Daily, Jeff↗
DOE OSTI2017

Distributed-Memory Breadth-First Search on Massive Graphs. IN: Parallel Graph Algorithms

This chapter studies the problem of traversing large graphs using the breadth-first search order on distributed-memory supercomputers. We consider both the traditional level-synchronous top-down algorithm as well as the recently discovered direction optimizing algorithm. We analyze the performance and scalability trade-offs in using different local data structures such as CSR and DCSC, enabling in-node multithreading, and graph decompositions such as 1D and 2D decomposition.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Considerations on the Use of Custom Accelerators for Big Data Analytics

Accelerators, including Graphic Processing Units (GPUs) for gen- eral purpose computation, many-core designs with wide vector units (e.g., Intel Phi), have become a common component of many high performance clusters. The appearance of more stable and reliable tools tools that can automatically convert code written in high-level specifications with annotations (such as C or C++) to hardware de- scription languages (High-Level Synthesis - HLS), is also setting the stage for a broader use of reconfigurable devices (e.g., Field Pro- grammable Gate Arrays - FPGAs) in high performance system for the implementation of custom accelerators, helped by the fact that new processors include advanced cache-coherent interconnects for these components. In this chapter, we briefly survey the status of the use of accelerators in high performance systems targeted at big data analytics applications. We argue that, although the progress in the use of accelerators for this class of applications has been sig- nificant, differently from scientific simulations there still are gaps to close. This is particularly true for the ”irregular” behaviors exhibited by no-SQL, graph databases. We focus our attention on the limits of HLS tools for data analytics and graph methods, and discuss a new architectural template that better fits the requirement of this class of applications. We validate the new architectural templates by mod- ifying the Graph Engine for Multithreaded System (GEMS) frame- work to support accelerators generated with such a methodology, and testing with queries coming from the Lehigh University Benchmark (LUBM). The architectural template enables better supporting the task and memory level parallelism present in graph methods by sup- porting a new control model and a enhanced memory interface. We show that out solution allows generating parallel accelerators, pro- viding speed ups with respect to conventional HLS flows. We finally draw conclusions and present a perspective on the use of reconfig- urable devices and Design Automation tools for data analytics.

Data Analytics↗
DOE OSTI2018

EnergyPlus Interior Radiant Heat Exchange Runtime Performance Improvements

EnergyPlus is the flagship whole-building energy simulation program developed by the US Department of Energy. This paper describes the various approaches to improve the performance of the interior radiant heat exchange algorithm. Vectorization, optimized Basic Linear Algebra Subprograms (BLAS) library, multithreading, and Graphical Processing Unit (GPU) computation are all investigated. The approach with the best performance, while maintaining drop-in compatibility with existing EnergyPlus code, is the optimized BLAS library. However, GPU computing has the potential to reduce simulation time by orders of magnitude with major refactoring to EnergyPlus.

Adams, Mark↗
DOE OSTI2019

Concurrent Conditions Access across Validity Intervals in CMSSW

The CMS software system, known as CMSSW, has a generalized conditions, calibration, and geometry data products system called the EventSetup. The EventSetup caches results of reading or calculating data products based on the 'interval of validity', IOV, which is based on the time period for which that data product is appropriate. With the original single threaded CMSSW framework, updating only on an IOV boundary meant we only required memory for a single data product of a given type at any time during the program execution. In 2016 CMS transitioned to using a multithreaded framework as a way to save on memory during processing. This was accomplished by amortizing the memory cost of EventSetup data products across multiple concurrent events. To initially accomplish that goal required synchronizing event processing across IOV boundaries, thereby decreasing the scalability of the system. In this presentation we will explain how we used 'limited concurrent task queues' to allow co ncurrent IOVs while still being able to limit the memory utilized. In addition, we will present performance measurements for both threading scalability of event throughtput and memory utilization.

Dagenhart, W.David↗
DOE OSTI2023

CMSSW Scaling Limits on Many-Core Machines

Today the LHC offline computing relies heavily on CPU resources, despite the interest in compute accelerators, such as GPUs, for the longer term future. The number of cores per CPU socket has continued to increase steadily, reaching the levels of 64 cores (128 threads) with recent AMD EPYC processors, and 128 cores on Ampere Altra Max ARM processors. Over the course of the past decade, the CMS data processing framework, CMSSW, has been transformed from a single-threaded framework into a highly concurrent one. The first multithreaded version was brought into production by the start of the LHC Run 2 in 2015. Since then, the framework's threading efficiency has gradually been improved by adding more levels of concurrency and reducing the amount of serial code paths. The latest addition was support for concurrent Runs. In this work we review the concurrency model of the CMSSW, and measure its scalability with real CMS applications, such as simulation and reconstruction, on mode rn many-core machines. We show metrics such as event processing throughput and application memory usage with and without the contribution of I/O, as I/O has been the major scaling limitation for the CMS applications.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2014

Parallel Monte Carlo Methods for Heterogeneous Hardware Computer Systems Using GPUs and Coprocessors: Recent Development of ARCHER Code - Paper 54

Traditional methods of accelerating Monte Carlo (MC) calculations typically use CPUs as the computing unit. The parallelism is based upon multithreading on a multi-core processor, and message passing interface (MPI) across many processors and nodes. In recent years, the hardware accelerators have been introduced by the high performance computing industry as an alternative approach to achieve massive parallelism. Hardware accelerators have the advantage of high Floating-point Operations Per Second (FLOPS) and high energy efficiency. Examples include graphics processing units (GPUs) and coprocessors. Compute-intensive programs can be off-loaded to them to be concurrently executed by hundreds or thousands of active threads. Existing production MC codes, however, cannot be directly compiled or executed on the new platforms. It is therefore necessary to rewrite or redevelop the code using the hardware accelerator specific programming models and APIs. Several studies have been conducted on using GPUs to accelerate MC calculations in reactor analysis and medical physics. In this paper, we present the recent progress at Rensselaer Polytechnic Institute (RPI) in developing a new parallel MC code called ARCHER (Accelerated Radiation-transport Computations in Heterogeneous Environments) for coupled photon-electron transport simulations. The code is designed as a test bed to evaluate the efficacy of heterogeneous computing systems for low-cost MC-based radiation dosimetric calculations. The code has three variants corresponding to three parallel hardware platforms: ARCHER-CPU (multi-core CPU), ARCHER-GPU (Nvidia GPU) and ARCHER-COP (Intel Xeon Phi coprocessor). The code has been used for several applications, including CT imaging dosimetry, radiotherapy dosimetry and radiation shielding design. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2014

Application of the Geant4 Simulation Toolkit to Medical Physics - Paper 65

We show how the Geant4 Simulation Toolkit, originally developed for High Energy Physics, has been applied to a wide variety of challenging applications in medical physics. As the authors are mainly responsible for providing assistance to Geant4 users in North America, they draw on examples from the North American experience, but the conclusions can be generalized to the Geant4 medical user community worldwide. Geant4 is already heavily used in all areas of medical physics, from external beam therapy to brachytherapy to interventional nuclear medicine and medical imaging. As the complexity of medical physics studies continues to increase, requiring greater accuracy of modeling, and an increase in 4D capabilities (component motion and time-varying electric and magnetic fields), Geant4 use will only expand. The openness of the HEP community (keeping Geant4 well supported, open and free) is greatly appreciated by the medical and aerospace communities. The democratic nature of Geant4's governance (that decision-making rests with a technical steering board elected by the collaborators) has given outside communities the confidence to invest their energies in Geant4 (shown them that it is worth the time investment to become expert users). Geant4's latest key area of focus, moving to multithreaded operation, is seen as a huge boon to the medical users (where precision simulations often require billions of primary particles). Another recent key area of Geant4, increasing automated systems for unit and end-to-end testing, is also seen as extremely valuable to these non-HEP communities. Innovations from HEP and aerospace will benefit medical applications, and vice versa. Geant4 can expect increasing resources to be provided by the medical physics community, pooling resources to the benefit of all users. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2026

EXCLUSIVE NEUTRAL PION ELECTROPRODUCTION CROSS SECTION MEASUREMENTSWITHANEUTRALPARTICLE SPECTROMETER

Deep Virtual Compton Scattering (DVCS), the exclusive electron-proton scattering process ep ¿e'p'¿, provides access to generalized parton distributions (GPDs), which correlate information about the longitudinal momentum and transverse spatial structure of quarks inside the nucleon. Experiment E12-13-010 in Hall C at Jefferson Lab was designed to take high-precision measurements of the DVCS cross section over an extended kinematic range using the newly commissioned Neutral Particle Spectrometer (NPS). The NPS features a high-resolution electromagnetic calorimeter and a streaming data acquisition system optimized for operation at high luminosities. This thesis presents the detector and analysis work carried out to support the NPS DVCS program. In particular, it focuses on the hardware design, calibration, and performance of the calorimeter. A development of a waveform reconstruction analysis of the calorimeter signals enabled improved extraction of pulse amplitudes and times. The waveform analysis was also extended to operate in a multithreaded environment, substantially reducing processing time for large datasets. Analysis of exclusive neutral pion electroproduction events in the calorimeter gives a strong validation of the calorimeter’s performance and resolution. Together these developments establish a foundation for future analyses and extraction of the DVCS cross section and its use in constraining the GPDs.

Kerver, Mitchell [Old Dominion Univ., Norfolk, VA ↗
DOE OSTI2014

Improved MPI collectives for MPI processes in shared address spaces

As the number of cores per node keeps increasing, it becomes increasingly important for MPI to leverage shared memory for intranode communication. This paper investigates the design and optimization of MPI collectives for clusters of NUMA nodes. We develop performance models for collective communication using shared memory and we demonstrate several algorithms for various collectives. Experiments are conducted on both Xeon X5650 and Opteron 6100 InfiniBand clusters. The measurements agree with the model and indicate that different algorithms dominate for short vectors and long vectors. We compare our shared-memory allreduce with several MPI implementations-Open MPI, MPICH2, and MVAPICH2-that utilize system shared memory to facilitate interprocess communication. On a 16-node Xeon cluster and 8-node Opteron cluster, our implementation achieves on geometric average 2.3X and 2.1X speedup over the best MPI implementation, respectively. Our techniques enable an efficient implementation of collective operations on future multi- and manycore systems.

Collective communication↗
DOE OSTI2021

Performance of Julia for High Energy Physics Analyses

We argue that the Julia programming language is a compelling alternative to currently more common implementations in Python and C++ for common data analysis workflows in high energy physics. We compare the speed of implementations of different workflows in Julia with those in Python and C++. Furthermore, our studies show that the Julia implementations are competitive for tasks that are dominated by computational load rather than data access. For work that is dominated by data access, we demonstrate an application with concurrent file reading and parallel data processing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

Toward a Data Scalable Solution for Facilitating Discovery of Science Resources

Science is increasingly motivated by the need to process larger quantities of data. It is facing severe challenges in data collection, management, and processing, so much so that the computational demands of “data scaling” are competing with, and in many fields surpassing, the traditional objective of decreasing processing time. Example domains with large datasets include astronomy, biology, genomics, climate/weather, and material sciences. This paper presents a real-world use case in which we wish to answer queries pro- vided by domain scientists in order to facilitate discovery of relevant science resources. The problem is that the metadata for these science resources is very large and is growing quickly, rapidly increasing the need for a data scaling solution. We propose a system – SGEM – designed for answering graph-based queries over large datasets on cluster architectures, and we re- port performance results for queries on the current RDESC dataset of nearly 1.4 billion triples, and on the well-known BSBM SPARQL query benchmark.

aggregation↗
DOE OSTI2017

A Locality-Based Threading Algorithm for the Configuration-Interaction Method

The Configuration Interaction (CI) method has been widely used to solve the non-relativistic many-body Schrodinger equation. One great challenge to implementing it efficiently on manycore architectures is its immense memory and data movement requirements. To address this issue, within each node, we exploit a hybrid MPI+OpenMP programming model in lieu of the traditional flat MPI programming model. Here in this paper, we develop optimizations that partition the workloads among OpenMP threads based on data locality,-which is essential in ensuring applications with complex data access patterns scale well on manycore architectures. The new algorithm scales to 256 threadson the 64-core Intel Knights Landing (KNL) manycore processor and 24 threads on dual-socket Ivy Bridge (Xeon) nodes. Compared with the original implementation, the performance has been improved by up to 7× on theKnights Landing processor and 3× on the dual-socket Ivy Bridge node.

97 MATHEMATICS AND COMPUTING↗