Search NASA⌕ Search

SEARCH · Search NASA

Results for “Coprocessor”

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 55 records · Page 3

An MPI + $X$ implementation of contact global search using Kokkos

This paper describes an approach that seeks to parallelize the spatial search associated with computational contact mechanics. In contact mechanics, the purpose of the spatial search is to find “nearest neighbors,” which is the prelude to an imprinting search that resolves the interactions between the external surfaces of contacting bodies. In particular, we are interested in the contact global search portion of the spatial search associated with this operation on domain-decomposition-based meshes. Specifically, we describe an implementation that combines standard domain-decomposition-based MPI-parallel spatial search with thread-level parallelism (MPI-X) available on advanced computer architectures (those with GPU coprocessors). Our goal is to demonstrate the efficacy of the MPI-X paradigm in the overall contact search. Standard MPI-parallel implementations typically use a domain decomposition of the external surfaces of bodies within the domain in an attempt to efficiently distribute computational work. This decomposition may or may not be the same as the volume decomposition associated with the host physics. The parallel contact global search phase is then employed to find and distribute surface entities (nodes and faces) that are needed to compute contact constraints between entities owned by different MPI ranks without further inter-rank communication. Key steps of the contact global search include computing bounding boxes, building surface entity (node and face) search trees and finding and distributing entities required to complete on-rank (local) spatial searches. To enable source-code portability and performance across a variety of different computer architectures, we implemented the algorithm using the Kokkos hardware abstraction library. While we targeted development towards machines with a GPU accelerator per MPI rank, we also report performance results for OpenMP with a conventional multi-core compute node per rank. Results here demonstrate a 47 % decrease in the time spent within the global search algorithm, comparing the reference ACME algorithm with the GPU implementation, on an 18M face problem using four MPI ranks. As a result, while further work remains to maximize performance on the GPU, this result illustrates the potential of the proposed implementation.

97 MATHEMATICS AND COMPUTING↗

Exploring the interoperability of remote GPGPU virtualization using rCUDA and directive-based programming models

Directive-based programming models, such as OpenMP, OpenACC, and OmpSs, enable users to accelerate applications by using coprocessors with little effort. These devices offer significant computing power, but their use can introduce two problems: an increase in the total cost of ownership and their underutilization because not all codes match their architecture. Remote accelerator virtualization frameworks address those problems. In particular, rCUDA provides transparent access to any graphic processor unit installed in a cluster, reducing the number of accelerators and increasing their utilization ratio. Joining these two technologies, directive-based programming models and rCUDA, is thus highly appealing. In this work, we study the integration of OmpSs and OpenACC with rCUDA, describing and analyzing several applications over three different hardware configurations that include two InfiniBand interconnections and three NVIDIA accelerators. Our evaluation reveals favorable performance results, showing low overhead and similar scaling factors when using remote accelerators instead of local devices.

97 MATHEMATICS AND COMPUTING↗

FPGA-Accelerated Machine Learning Inference as a Service for Particle Physics Computing

Large-scale particle physics experiments face challenging demands for high-throughput computing resources both now and in the future. New heterogeneous computing paradigms on dedicated hardware with increased parallelization, such as Field Programmable Gate Arrays (FPGAs), offer exciting solutions with large potential gains. The growing applications of machine learning algorithms in particle physics for simulation, reconstruction, and analysis are naturally deployed on such platforms. We demonstrate that the acceleration of machine learning inference as a web service represents a heterogeneous computing solution for particle physics experiments that potentially requires minimal modification to the current computing model. As examples, we retrain the ResNet-50 convolutional neural network to demonstrate state-of-the-art performance for top quark jet tagging at the LHC and apply a ResNet-50 model with transfer learning for neutrino event classification. Using Project Brainwave by Microsoft to accelerate the ResNet-50 image classification model, we achieve average inference times of 60 (10) ms with our experimental physics software framework using Brainwave as a cloud (edge or on-premises) service, representing an improvement by a factor of approximately 30 (175) in model inference latency over traditional CPU inference in current experimental hardware. A single FPGA service accessed by many CPUs achieves a throughput of 600–700 inferences per second using an image batch of one, comparable to large batch-size GPU throughput and significantly better than small batch-size GPU throughput. Deployed as an edge or cloud service for the particle physics computing model, coprocessor accelerators can have a higher duty cycle and are potentially much more cost-effective.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Implementing Molecular Dynamics on Hybrid High Performance Computers - Three-Body Potentials

The use of coprocessors or accelerators such as graphics processing units (GPUs) has become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power re- quirements. Hybrid high-performance computers, defined as machines with nodes containing more than one type of floating-point processor (e.g. CPU and GPU), are now becoming more prevalent due to these advantages. Although there has been extensive research into methods to efficiently use accelerators to improve the performance of molecular dynamics (MD) employing pairwise potential energy models, little is reported in the literature for models that include many-body effects. 3-body terms are required for many popular potentials such as MEAM, Tersoff, REBO, AIREBO, Stillinger-Weber, Bond-Order Potentials, and others. Because the per-atom simulation times are much higher for models incorporating 3-body terms, there is a clear need for efficient algo- rithms usable on hybrid high performance computers. Here, we report a shared-memory force-decomposition for 3-body potentials that avoids memory conflicts to allow for a deterministic code with substantial performance improvements on hybrid machines. We describe modifications necessary for use in distributed memory MD codes and show results for the simulation of water with Stillinger-Weber on the hybrid Titan supercomputer. We compare performance of the 3-body model to the SPC/E water model when using accelerators. Finally, we demonstrate that our approach can attain a speedup of 5.1 with acceleration on Titan for production simulations to study water droplet freezing on a surface.

Brown, W Michael↗

The high level trigger and express data production at STAR

To meet the demands of the Beam Energy Scan phase-II (BES-II) program, the STAR experiment at the Relativistic Heavy Ion Collider (RHIC) developed a dual real-time framework consisting of a High Level Trigger (HLT) and an Express Data Production system (xProduction). The HLT operates online within the Data Acquisition (DAQ) chain on a dedicated multi-core CPU cluster with the option to offload compute-intensive kernels to Xeon Phi coprocessors. It uses parallelized algorithms, such as the Cellular Automaton (CA) Track Finder, to perform rapid tracking, vertexing, and event filtering. This allows it to select events of interest in real time and provide immediate feedback on detector and beam conditions. In contrast, the xProduction workflow runs concurrently and independently of the DAQ loop. It applies near offline-quality calibration and reconstruction within hours of data collection. The xProduction input is the express data stream, whose content can be enriched by HLT trigger/priority selections under DAQ/HLT resource constraints, and it uses the STAR calibration/conditions framework, incorporating online calibration/QA information when available. This enables early preliminary physics analysis, including the reconstruction of rare signals, such as hyperons and hypernuclei. It also provides collaboration-wide access to analysis-ready datasets. Together, the HLT and xProduction systems form a complementary architecture: the HLT performs online event selection while the xProduction chain delivers high-quality results within a short amount of time. This integrated framework has enabled the prompt reconstruction of the $^5_Λ$ He hypernucleus with high statistical significance and the efficient processing of hundreds of millions of heavy-ion collision events. In conclusion, its demonstrated scalability and robustness establish a model for future high-luminosity experiments requiring both online event filtering and rapid access to analysis-quality data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Impact of detector simulation in particle physics collider experiments

Through the last three decades, precise simulation of the interactions of particles with matter and modeling of detector geometries has proven to be of critical importance to the success of the international high-energy physics experimental programs. For example, the detailed detector modeling and accurate physics of the Geant4-based simulation software of the CMS and ATLAS particle physics experiments at the European Center of Nuclear Research (CERN) Large Hadron Collider (LHC) was a determinant factor for these collaborations to deliver physics results of outstanding quality faster than any hadron collider experiment ever before. This review article highlights the impact of detector simulation on particle physics collider experiments. It presents numerous examples of the use of simulation, from detector design and optimization, through software and computing development and testing, to cases where the use of simulation samples made a difference in the accuracy of the physics results and publication turnaround, from data-taking to submission. It also presents the economic impact and cost of simulation in the CMS experiment. Future experiments will collect orders of magnitude more data, taxing heavily the performance of simulation and reconstruction software for increasingly complex detectors. Consequently, it becomes urgent to find solutions to speed up simulation software in order to cope with the increased demand in a time of flat budgets. The study ends with a short discussion on the potential solutions that are being explored, by leveraging core count growth in multicore machines, using new generation coprocessors, and re-engineering of HEP code for concurrency and parallel computing.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

ATLAS FTK Challenge: Simulation of a Billion-fold Hardware Parallelism

During the current LHC shutdown period the ATLAS experiment will upgrade the Trigger and Data Acquisition system to include a hardware tracker coprocessor: the Fast TracKer (FTK). The FTK receives data from the 80 million of channels of the ATLAS silicon detector, identifying charged tracks and reconstructing their parameters at a rate of up to 100 KHz and within 100 microseconds. To achieve this performance, the FTK system identifies candidate tracks utilizing the computing power of a custom ASIC chip with associative memory (AM) designed to perform “pattern matching” at very high speed; track parameters are then calculated using modern FPGAs. A detailed simulation of this massive system has been developed with the goal of supporting the hardware design and studying its impact in the ATLAS online event selection at high LHC luminosities. We present the issues related to emulating FTK on a general-purpose CPU platform, using ATLAS computing Grid resources, and the solutions developed in order to mitigate these problems and allow the studies required to support the system design, construction and installation.

Distributed Computing↗

VINE-A NUMERICAL CODE FOR SIMULATING ASTROPHYSICAL SYSTEMS USING PARTICLES. I. DESCRIPTION OF THE PHYSICS AND THE NUMERICAL METHODS

We present a numerical code for simulating the evolution of astrophysical systems using particles to represent the underlying fluid flow. The code is written in Fortran 95 and is designed to be versatile, flexible, and extensible, with modular options that can be selected either at the time the code is compiled or at run time through a text input file. We include a number of general purpose modules describing a variety of physical processes commonly required in the astrophysical community and we expect that the effort required to integrate additional or alternate modules into the code will be small. In its simplest form the code can evolve the dynamical trajectories of a set of particles in two or three dimensions using a module which implements either a Leapfrog or Runge-Kutta-Fehlberg integrator, selected by the user at compile time. The user may choose to allow the integrator to evolve the system using individual time steps for each particle or with a single, global time step for all. Particles may interact gravitationally as N-body particles, and all or any subset may also interact hydrodynamically, using the smoothed particle hydrodynamic (SPH) method by selecting the SPH module. A third particle species can be included with a module to model massive point particles which may accrete nearby SPH or N-body particles. Such particles may be used to model, e.g., stars in a molecular cloud. Free boundary conditions are implemented by default, and a module may be selected to include periodic boundary conditions. We use a binary 'Press' tree to organize particles for rapid access in gravity and SPH calculations. Modules implementing an interface with special purpose 'GRAPE' hardware may also be selected to accelerate the gravity calculations. If available, forces obtained from the GRAPE coprocessors may be transparently substituted for those obtained from the tree, or both tree and GRAPE may be used as a combination GRAPE/tree code. The code may be run without modification on single processors or in parallel using OpenMP compiler directives on large-scale, shared memory parallel machines. We present simulations of several test problems, including a merger simulation of two elliptical galaxies with 800,000 particles. In comparison to the Gadget-2 code of Springel, the gravitational force calculation, which is the most costly part of any simulation including self-gravity, is {approx}4.6-4.9 times faster with VINE when tested on different snapshots of the elliptical galaxy merger simulation when run on an Itanium 2 processor in an SGI Altix. A full simulation of the same setup with eight processors is a factor of 2.91 faster with VINE. The code is available to the public under the terms of the Gnu General Public License.

79 ASTRONOMY AND ASTROPHYSICS↗

First experience of vectorizing electromagnetic physics models for detector simulation

The recent emergence of hardware architectures characterized by many-core or accelerated processors has opened new opportunities for concurrent programming models taking advantage of both SIMD and SIMT architectures. The GeantV vector prototype for detector simulations has been designed to exploit both the vector capability of mainstream CPUs and multi-threading capabilities of coprocessors including NVidia GPUs and Intel Xeon Phi. The characteristics of these architectures are very different in terms of the vectorization depth, parallelization needed to achieve optimal performance or memory access latency and speed. An additional challenge is to avoid the code duplication often inherent to supporting heterogeneous platforms. In this paper we present the first experience of vectorizing electromagnetic physics models developed for the GeantV project.

Amadio, G.↗

Electromagnetic physics models for parallel computing architectures

The recent emergence of hardware architectures characterized by many-core or accelerated processors has opened new opportunities for concurrent programming models taking advantage of both SIMD and SIMT architectures. GeantV, a next generation detector simulation, has been designed to exploit both the vector capability of mainstream CPUs and multi-threading capabilities of coprocessors including NVidia GPUs and Intel Xeon Phi. The characteristics of these architectures are very different in terms of the vectorization depth and type of parallelization needed to achieve optimal performance. In this paper we describe implementation of electromagnetic physics models developed for parallel computing architectures as a part of the GeantV project. Finally, the results of preliminary performance evaluation and physics validation are presented as well.

97 MATHEMATICS AND COMPUTING↗

Performance of GeantV EM Physics Models

The recent progress in parallel hardware architectures with deeper vector pipelines or many-cores technologies brings opportunities for HEP experiments to take advantage of SIMD and SIMT computing models. Launched in 2013, the GeantV project studies performance gains in propagating multiple particles in parallel, improving instruction throughput and data locality in HEP event simulation on modern parallel hardware architecture. Due to the complexity of geometry description and physics algorithms of a typical HEP application, performance analysis is indispensable in identifying factors limiting parallel execution. In this report, we will present design considerations and preliminary computing performance of GeantV physics models on coprocessors (Intel Xeon Phi and NVidia GPUs) as well as on mainstream CPUs.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

MIC-SVM: Designing A Highly Efficient Support Vector Machine For Advanced Modern Multi-Core and Many-Core Architectures

Support Vector Machine (SVM) has been widely used in data-mining and Big Data applications as modern commercial databases start to attach an increasing importance to the analytic capabilities. In recent years, SVM was adapted to the field of High Performance Computing for power/performance prediction, auto-tuning, and runtime scheduling. However, even at the risk of losing prediction accuracy due to insufficient runtime information, researchers can only afford to apply offline model training to avoid significant runtime training overhead. To address the challenges above, we designed and implemented MICSVM, a highly efficient parallel SVM for x86 based multi-core and many core architectures, such as the Intel Ivy Bridge CPUs and Intel Xeon Phi coprocessor (MIC).

42 ENGINEERING↗

Multi-threaded Geant4 on the Xeon-Phi with complex high-energy physics geometry

To study the performance of multi-threaded Geant4 for high-energy physics experiments, an application has been developed which generalizes and extends previous work. Here, a highly-complex detector geometry is used for benchmarking on an Intel Xeon Phi coprocessor. Additonally, an implementation of parallel I/O based on Intel SCIF and ROOT technologies is incorporated and studied.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Computing Maximum Cardinality Matchings in Parallel on Bipartite Graphs via Tree-Grafting

It is difficult to obtain high performance when computing matchings on parallel processors because matching algorithms explicitly or implicitly search for paths in the graph, and when these paths become long, there is little concurrency. In spite of this limitation, we present a new algorithm and its shared-memory parallelization that achieves good performance and scalability in computing maximum cardinality matchings in bipartite graphs. This algorithm searches for augmenting paths via specialized breadth-first searches (BFS) from multiple source vertices, hence creating more parallelism than single source algorithms. Algorithms that employ multiple-source searches cannot discard a search tree once no augmenting path is discovered from the tree, unlike algorithms that rely on single-source searches. We describe a novel tree-grafting method that eliminates most of the redundant edge traversals resulting from this property of multiple-source searches. We also employ the recent direction-optimizing BFS algorithm as a subroutine to discover augmenting paths faster. Our algorithm compares favorably with the current best algorithms in terms of the number of edges traversed, the average augmenting path length, and the number of iterations. Here, we provide a proof of correctness for our algorithm. Our NUMA-aware implementation is scalable to 80 threads of an Intel multiprocessor and to 240 threads on an Intel Knights Corner coprocessor. On average, our parallel algorithm runs an order of magnitude faster than the fastest algorithms available. The performance improvement is more significant on graphs with small matching number.

97 MATHEMATICS AND COMPUTING↗

Probing Quantum Telecloning on Superconducting Quantum Processors

Quantum information cannot be perfectly cloned, but approximate copies of quantum information can be generated. Quantum telecloning combines approximate quantum cloning, more typically referred to as quantum cloning, and quantum teleportation. Quantum telecloning allows approximate copies of quantum information to be constructed by separate parties, using the classical results of a Bell measurement made on a prepared quantum telecloning state. Quantum telecloning can be implemented as a circuit on quantum computers using a classical coprocessor to compute classical feedforward instructions using if statements based on the results of a midcircuit Bell measurement in real time. We present universal symmetric optimal 1 → M telecloning circuits and experimentally demonstrate these quantum telecloning circuits for M = 2 up to M = 10, natively executed with real-time classical control systems on IBM Quantum superconducting processors, known as dynamic circuits. We perform the cloning procedure on many different message states across the Bloch sphere, on seven IBM Quantum processors, optionally using the error suppression technique X–X sequence digital dynamical decoupling. Two circuit optimizations are utilized: one that removes ancilla qubits for M = 2,3 , and one that reduces the total number of gates in the circuit but still uses ancilla qubits. Parallel single-qubit tomography with maximum likelihood estimation density matrix reconstruction is used in order to compute the mixed-state density matrices of the clone qubits, and clone quality is measured using quantum fidelity. These results present one of the largest and most comprehensive noisy intermediate-scale quantum computer experimental analyses on (single qubit) quantum telecloning to date. The clone fidelity sharply decreases to 0.5 for M > 5 , but for M = 2 , we are able to achieve a mean clone fidelity of up to 0.79 using dynamical decoupling.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Towards Achieving Performance Portability Using Directives for Accelerators

In this paper we explore the performance portability of directives provided by OpenMP 4 and OpenACC to program various types of node architectures with attached accelerators, both self-hosted multicore and offload multicore/GPU. Our goal is to examine how successful OpenACC and the newer offload features of OpenMP 4.5 are for moving codes between architectures, how much tuning might be required and what lessons we can learn from this experience. To do this, we use examples of algorithms with varying computational intensities for our evaluation, as both compute and data access efficiency are important considerations for overall application performance. We implement these kernels using various methods provided by newer OpenACC and OpenMP implementations, and we evaluate their performance on various platforms including both X86_64 with attached NVIDIA GPUs, self-hosted Intel Xeon Phi KNL, as well as an X86_64 host system with Intel Xeon Phi coprocessors. In this paper, we explain what factors affected the performance portability such as how to pick the right programming model, its programming style, its availability on different platforms, and how well compilers can optimize and target to multiple platforms.

97 MATHEMATICS AND COMPUTING↗

TH-A-19A-08: Intel Xeon Phi Implementation of a Fast Multi-Purpose Monte Carlo Simulation for Proton Therapy

Purpose: Recent studies have demonstrated the capability of graphics processing units (GPUs) to compute dose distributions using Monte Carlo (MC) methods within clinical time constraints. However, GPUs have a rigid vectorial architecture that favors the implementation of simplified particle transport algorithms, adapted to specific tasks. Our new, fast, and multipurpose MC code, named MCsquare, runs on Intel Xeon Phi coprocessors. This technology offers 60 independent cores, and therefore more flexibility to implement fast and yet generic MC functionalities, such as prompt gamma simulations. Methods: MCsquare implements several models and hence allows users to make their own tradeoff between speed and accuracy. A 200 MeV proton beam is simulated in a heterogeneous phantom using Geant4 and two configurations of MCsquare. The first one is the most conservative and accurate. The method of fictitious interactions handles the interfaces and secondary charged particles emitted in nuclear interactions are fully simulated. The second, faster configuration simplifies interface crossings and simulates only secondary protons after nuclear interaction events. Integral depth-dose and transversal profiles are compared to those of Geant4. Moreover, the production profile of prompt gammas is compared to PENH results. Results: Integral depth dose and transversal profiles computed by MCsquare and Geant4 are within 3%. The production of secondaries from nuclear interactions is slightly inaccurate at interfaces for the fastest configuration of MCsquare but this is unlikely to have any clinical impact. The computation time varies between 90 seconds for the most conservative settings to merely 59 seconds in the fastest configuration. Finally prompt gamma profiles are also in very good agreement with PENH results. Conclusion: Our new, fast, and multi-purpose Monte Carlo code simulates prompt gammas and calculates dose distributions in less than a minute, which complies with clinical time constraints. It has been successfully validated with Geant4. This work has been financialy supported by InVivoIGT, a public/private partnership between UCL and IBA.

07 ISOTOPE AND RADIATION SOURCES↗

TU-AB-BRC-10: Modeling of Radiotherapy Linac Source Terms Using ARCHER Monte Carlo Code: Performance Comparison of GPU and MIC Computing Accelerators

Purpose: (1) To perform phase space (PS) based source modeling for Tomotherapy and Varian TrueBeam 6 MV Linacs, (2) to examine the accuracy and performance of the ARCHER Monte Carlo code on a heterogeneous computing platform with Many Integrated Core coprocessors (MIC, aka Xeon Phi) and GPUs, and (3) to explore the software micro-optimization methods. Methods: The patient-specific source of Tomotherapy and Varian TrueBeam Linacs was modeled using the PS approach. For the helical Tomotherapy case, the PS data were calculated in our previous study (Su et al. 2014 41(7) Medical Physics). For the single-view Varian TrueBeam case, we analytically derived them from the raw patient-independent PS data in IAEA’s database, partial geometry information of the jaw and MLC as well as the fluence map. The phantom was generated from DICOM images. The Monte Carlo simulation was performed by ARCHER-MIC and GPU codes, which were benchmarked against a modified parallel DPM code. Software micro-optimization was systematically conducted, and was focused on SIMD vectorization of tight for-loops and data prefetch, with the ultimate goal of increasing 512-bit register utilization and reducing memory access latency. Results: Dose calculation was performed for two clinical cases, a Tomotherapy-based prostate cancer treatment and a TrueBeam-based left breast treatment. ARCHER was verified against the DPM code. The statistical uncertainty of the dose to the PTV was less than 1%. Using double-precision, the total wall time of the multithreaded CPU code on a X5650 CPU was 339 seconds for the Tomotherapy case and 131 seconds for the TrueBeam, while on 3 5110P MICs it was reduced to 79 and 59 seconds, respectively. The single-precision GPU code on a K40 GPU took 45 seconds for the Tomotherapy dose calculation. Conclusion: We have extended ARCHER, the MIC and GPU-based Monte Carlo dose engine to Tomotherapy and Truebeam dose calculations.

60 APPLIED LIFE SCIENCES↗