Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “execution time”

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 145 records · Page 8

Building Scalable PGAS Communication Subsystem on Blue Gene/Q

This paper presents a design of scalable Partitioned Global Address Space (PGAS) communication subsystems on recently proposed Blue Gene/Q architecture. The proposed design provides an in-depth modeling of communication infrastructure using Parallel Active Messaging Interface (PAMI). The communication infrastructure is used to design time-space efficient communication protocols for frequently used data-types (contiguous, uniformly non-contiguous) using Remote Direct Memory Access (RDMA) get/put primitives. The proposed design accelerates load balance counters by using asynchronous threads, which are required due to the missing network hardware support for Atomic Memory Operations (AMOs). Under the proposed design, the synchronization traffic is reduced by tracking conflicting memory accesses in distributed space with slightly increment in space complexity. An evaluation with simple communication benchmarks show a adjacent node get latency of 2.89$\mu$s and peak bandwidth of 1775 MB/s resulting in $\approx$ 99\% communication efficiency.The evaluation shows a reduction in the execution time by up to 30\% for NWChem self consistent field calculation on 4096 processes using the proposed asynchronous thread based design.

Vishnu, Abhinav↗

Quantum Computing in the Cloud: Analyzing job and machine characteristics

As the popularity of quantum computing continues to grow, quantum machine access over the cloud is critical to both academic and industry researchers across the globe. And as cloud quantum computing demands increase exponentially, the analysis of resource consumption and execution characteristics are key to efficient management of jobs and resources at both the vendor-end as well as the client-end. While the analysis of resource consumption and management are popular in the classical HPC domain, it is severely lacking for more nascent technology like quantum computing. This paper is a first-of-its-kind academic study, analyzing various trends in job execution and resources consumption / utilization on quantum cloud systems. We focus on IBM Furthermore, quantum systems and analyze characteristics over a two year period, encompassing over 6000 jobs which contain over 600,000 quantum circuit executions and correspond to almost 10 billion “shots” or trials over 20+ quantum machines. Specifically, we analyze trends focused on, but not limited to, execution times on quantum machines, queuing/waiting times in the cloud, circuit compilation times, machine utilization, as well as the impact of job and machine characteristics on all of these trends. Furthermore, our analysis identifies several similarities and differences with classical HPC cloud systems. Based on our insights, we make recommendations and contributions to improve the management of resources and jobs on future quantum cloud systems.

42 ENGINEERING↗

Cooperative Server Clustering for a Scalable GAS Model on Petascale Cray XT5 Systems

Global Address Space (GAS) programming models are attractive because they retain the easy-to-use addressing model that is the characteristic of shared-memory style load and store operations. The scalability of GAS models depends directly on the design and implementation of runtime libraries on the targeted platforms. In this paper, we examine the memory requirement of a popular GAS run-time library, Aggregate Remote Memory Copy Interface (ARMCI) on petascale Cray XT 5 systems. Then we describe a new technique, cooperative server clustering, that enhances the memory scalability of ARMCI communication servers. In cooperative server clustering, ARMCI servers are organized into clusters, and cooperatively process incoming communication requests among them. A request intervention scheme is also designed to expedite the return of responses to the initiating processes. Our experimental results demonstrate that, with very little impact on ARMCI communication latency and bandwidth, cooperative server clustering is able to significantly reduce the memory requirement of ARMCI communication servers, thereby enabling highly scalable scientific applications. In particular, it dramatically reduces the total execution time of a scientific application, NWChem, by 45% on 2400 processes.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Demonstrating the viability of Lagrangian in situ reduction on supercomputers

Performing exploratory analysis and visualization of large-scale time-varying computational science applications is challenging due to inaccuracies that arise from under-resolved data. In recent years, Lagrangian representations of the vector field computed using in situ processing are being increasingly researched and have emerged as a potential solution to enable exploration. However, prior works have offered limited estimates of the encumbrance on the simulation code as they consider “theoretical” in situ environments. Further, the effectiveness of this approach varies based on the nature of the vector field, benefitting from an in-depth investigation for each application area. With this study, an extended version of Sane et al. (2021), we contribute an evaluation of Lagrangian analysis viability and efficacy for simulation codes executing at scale on a supercomputer. We investigated previously unexplored cosmology and seismology applications as well as conducted a performance benchmarking study by using a hydrodynamics mini-application targeting exascale computing. Here, to inform encumbrance, we integrated in situ infrastructure with simulation codes, and evaluated Lagrangian in situ reduction in representative homogeneous and heterogeneous HPC environments. To inform post hoc accuracy, we conducted a statistical analysis across a range of spatiotemporal configurations as well as a qualitative evaluation. Additionally, our study contributes cost estimates for distributed-memory post hoc reconstruction. In all, we demonstrate viability for each application — data reduction to less than 1% of the total data via Lagrangian representations, while maintaining accurate reconstruction and requiring under 10% of total execution time in over 90% of our experiments.

97 MATHEMATICS AND COMPUTING↗

Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive-Based Methods

Cloud microphysics is one of the most time-consuming components in a climate model. In this study, we port the cloud microphysics parameterization in the Community Atmosphere Model (CAM), known as Parameterization of Unified Microphysics Across Scales (PUMAS), from CPU to GPU to seek a computational speedup. The directive-based methods (OpenACC and OpenMP target offload) are determined as the best fit specifically for our development practices, which enable a single version of source code to run either on the CPU or GPU, and yield a better portability and maintainability. Their performance is first examined in a PUMAS stand-alone kernel and the directive-based methods can outperform a CPU node as long as there is enough computational burden on the GPU. A consistent behavior is observed when we run PUMAS on the GPU in a practical CAM simulation. A 3.6× speedup of the PUMAS execution time, including data movement between CPU and GPU, is achieved at a coarse horizontal resolution (8 NVIDIA V100 GPUs against 36 Intel Skylake CPU cores). This speedup further increases up to 5.4× at a high resolution (24 NVIDIA V100 GPUs against 108 Intel Skylake CPU cores), which highlights the fact that GPU favors larger problem size. This study demonstrates that using GPU in a CAM simulation can save noticeable computational costs even with a small portion of code being GPU-enabled. Therefore, we are encouraged to port more parameterizations to GPU to take advantage of its computational benefit.

54 ENVIRONMENTAL SCIENCES↗

Expediting Scientific Data Analysis with Reorganization of Data

Data producers typically optimize the layout of data files to minimize the write time. In most cases, data analysis tasks read these files in access patterns different from the write patterns causing poor read performance. In this paper, we introduce Scientific Data Services (SDS), a framework for bridging the performance gap between writing and reading scientific data. SDS reorganizes data to match the read patterns of analysis tasks and enables transparent data reads from the reorganized data. We implemented a HDF5 Virtual Object Layer (VOL) plugin to redirect the HDF5 dataset read calls to the reorganized data. To demonstrate the effectiveness of SDS, we applied two parallel data organization techniques: a sort-based organization on a plasma physics data and a transpose-based organization on mass spectrometry imaging data. We also extended the HDF5 data access API to allow selection of data based on their values through a query interface, called SDS Query. We evaluated the execution time in accessing various subsets of data through existing HDF5 Read API and SDS Query. We showed that reading the reorganized data using SDS is up to 55X faster than reading the original data.

97 MATHEMATICS AND COMPUTING↗

Monte Carlo simulator of realistic x-ray beam for diagnostic applications

Purpose: Monte Carlo simulation is a very useful tool for radiotherapy and diagnostic radiology. Yet even with the latest PCs, simulation of photon spectra emitted by an x-ray tube is a time-consuming task, potentially reducing the possibility to obtain relevant data such as dose evaluations, simulation of geometric settings, or monitor detector efficiency. This study developed and validated a method to generate random numbers for realistic beams in terms of photon spectrum and intensity to simulate x-ray tubes via Monte Carlo algorithms. Methods: Starting from literature data, the most common semiempirical models of bremsstrahlung are analyzed and implemented, adjusting their formulation to describe a large irradiation area (i.e., large field of view) and to take account of the heel effect as in common practice during patient examinations. Results: Simulation results show that Birch and Marshall's model is the fastest and most accurate for the aims of this work. Correction of the geometric size of the beam and validation of the intensity variation (heel effect) yielded excellent results with differences between experimental and simulated data of less than 6%. Conclusions: The results of validation and execution time showed that the tube simulator calculates the x-ray photons quickly and efficiently and is perfectly capable of considering all the phenomena occurring in a real beam (total filtration, focal spot size, and heel effect), so it can be used in a wide range of applications such as industry, medical physics, or quality assurance.

60 APPLIED LIFE SCIENCES↗

Optimizing Sparse Matrix-Multiple Vectors Multiplication for Nuclear Configuration Interaction Calculations

Obtaining highly accurate predictions on the properties of light atomic nuclei using the configuration interaction (CI) approach requires computing a few extremal Eigen pairs of the many-body nuclear Hamiltonian matrix. In the Many-body Fermion Dynamics for nuclei (MFDn) code, a block Eigen solver is used for this purpose. Due to the large size of the sparse matrices involved, a significant fraction of the time spent on the Eigen value computations is associated with the multiplication of a sparse matrix (and the transpose of that matrix) with multiple vectors (SpMM and SpMM-T). Existing implementations of SpMM and SpMM-T significantly underperform expectations. Thus, in this paper, we present and analyze optimized implementations of SpMM and SpMM-T. We base our implementation on the compressed sparse blocks (CSB) matrix format and target systems with multi-core architectures. We develop a performance model that allows us to understand and estimate the performance characteristics of our SpMM kernel implementations, and demonstrate the efficiency of our implementation on a series of real-world matrices extracted from MFDn. In particular, we obtain 3-4 speedup on the requisite operations over good implementations based on the commonly used compressed sparse row (CSR) matrix format. The improvements in the SpMM kernel suggest we may attain roughly a 40% speed up in the overall execution time of the block Eigen solver used in MFDn.

97 MATHEMATICS AND COMPUTING↗

Quantitative Performance Assessment of Proxy Apps and Parents (Report for ECP Proxy App Project Milestone ADCD-504-28)

The ECP Proxy Application Project has an annual milestone to assess the state of ECP proxy applications and their role in the overall ECP ecosystem. Our FY22 March/April milestone (ADCD- 504-28) proposed to: Assess the fidelity of proxy applications compared to their respective parents in terms of kernel and I/O behavior, and predictability. Similarity techniques will be applied for quantitative comparison of proxy/parent kernel behavior. MACSio evaluation will continue and support for OpenPMD backends will be explored. The execution time predictability of proxy apps with respect to their parents will be explored through a carefully designed scaling study and code comparisons. Note that in this FY, we also have quantitative assessment milestones that are due in September and are, therefore, not included in the description above or in this report. Another report on these deliverables will be generated and submitted upon completion of these milestones. To satisfy this milestone, the following specific tasks were completed: Study the ability of MACSio to represent I/O workloads of adaptive mesh codes. Re-define the performance counter groups for contemporary Intel and IBM platforms to better match specific hardware components and to better align across platforms (make cross-platform comparison more accurate). Perform cosine similarity study based on the new performance counter groups on the Intel and IBM P9 platforms. Perform detailed analysis of performance counter data to accurately average and align the data to maintain phases across all executions and develop methods to reduce the set of collected performance counters used in cosine similarity analysis. Apply a quantitative similarity comparison between proxy and parent CPU kernels. Perform scaling studies to understand the accuracy of predictability of the parent performance using its respective proxy application. This report presents highlights of these efforts.

97 MATHEMATICS AND COMPUTING↗

Implementation of the D1S Methodology for Shutdown Dose Rate Calculations in the OpenMC Monte Carlo Particle Transport Code

We present an implementation of the direct one-step (D1S) methodology for shutdown dose rate (SDR) calculations in the OpenMC Monte Carlo particle transport code. In addition to being the first fully open-source D1S implementation, it is also the first to require no ad hoc source code or nuclear data library modifications. The code can seamlessly switch between production of prompt and decay photons based on a user input parameter, and the decay data needed for decay photon generation are made available through a depletion chain file, which is already used for OpenMC’s built-in depletion/activation solver. A set of Python functions significantly eases the burden of computing and applying time correction factors needed to properly account for the time dependence of radionuclide activity. To assess the accuracy of the D1S implementation, SDR calculations have been carried out for three problems: a prism of iron irradiated by 14-MeV neutrons, the ITER port plug computational benchmark, and the Frascati Neutron Generator (FNG) ITER dose rate benchmark problem from the Shielding INtegral Benchmark Archive and Database (SINBAD). For each of these problems, comparisons were made to calculations using the rigorous two-step (R2S) method. The results on the iron prism problem illustrate how the D1S method achieves superior spatial resolution compared to the R2S method without the need for spatial discretization of the activation regions. The D1S and R2S results for the ITER port plug benchmark agree well with previously reported results in the literature. While the D1S results are 10% to 15% lower than the R2S results, this may be due to stochastic uncertainty and/or spatial discretization in the R2S calculations. On the FNG dose rate benchmark problem, the D1S method produces dose rate estimates that are within 4% of the dose rates predicted using a cell-based R2S workflow. The D1S estimates of the SDR are also in reasonable agreement with the experimental measurements and show the same basic trends that have been observed in previous works. A qualitative analysis of the execution time and uncertainty for the R2S and D1S workflows suggests that the D1S method would attain a higher figure of merit.

D1S method↗

Towards Enhancing Coding Productivity for GPU Programming Using Static Graphs

The main contribution of this work is to increase the coding productivity of GPU programming by using the concept of Static Graphs. GPU capabilities have been increasing significantly in terms of performance and memory capacity. However, there are still some problems in terms of scalability and limitations to the amount of work that a GPU can perform at a time. To minimize the overhead associated with the launch of GPU kernels, as well as to maximize the use of GPU capacity, we have combined the new CUDA Graph API with the CUDA programming model (including CUDA math libraries) and the OpenACC programming model. We use as test cases two different, well-known and widely used problems in HPC and AI: the Conjugate Gradient method and the Particle Swarm Optimization. In the first test case (Conjugate Gradient) we focus on the integration of Static Graphs with CUDA. In this case, we are able to significantly outperform the NVIDIA reference code, reaching an acceleration of up to 11x thanks to a better implementation, which can benefit from the new CUDA Graph capabilities. In the second test case (Particle Swarm Optimization), we complement the OpenACC functionality with the use of CUDA Graph, achieving again accelerations of up to one order of magnitude, with average speedups ranging from 2x to 4x, and performance very close to a reference and optimized CUDA code. Our main target is to achieve a higher coding productivity model for GPU programming by using Static Graphs, which provides, in a very transparent way, a better exploitation of the GPU capacity. The combination of using Static Graphs with two of the current most important GPU programming models (CUDA and OpenACC) is able to reduce considerably the execution time w.r.t. the use of CUDA and OpenACC only, achieving accelerations of up to more than one order of magnitude. Finally, we propose an interface to incorporate the concept of Static Graphs into the OpenACC Specifications.

58 GEOSCIENCES↗

An Efficient Algorithm for Mapping Imaging Data to 3D Unstructured Grids in Computational Biomechanics

Geometries for organ scale and multiscale simulations of organ function are now routinely derived from imaging data. However, medical images may also contain spatially heterogeneous information other than geometry that are relevant to such simulations either as initial conditions or in the form of model parameters. In this manuscript, we present an algorithm for the efficient and robust mapping of such data to imaging based unstructured polyhedral grids in parallel. We then illustrate the application of our mapping algorithm to three different mapping problems: 1) the mapping of MRI diffusion tensor data to an unstuctured ventricular grid; 2) the mapping of serial cyro-section histology data to an unstructured mouse brain grid; and 3) the mapping of CT-derived volumetric strain data to an unstructured multiscale lung grid. Execution times and parallel performance are reported for each case.

Einstein, Daniel R.↗

Multi‐dimensional lock‐free arrays for multithreaded mode‐directed tabling in Prolog

Summary This work proposes a new design for the supporting data structures used to implement multithreaded tabling in Prolog systems. Tabling is an implementation technique that improves the expressiveness of traditional Prolog systems in dealing with recursion and redundant computations. Mode‐directed tabling is an extension to the tabling technique that supports the definition of alternative criteria for specifying how answers are aggregated, thus being very suitable for problems where the goal is to dynamically calculate optimal or selective answers. In this work, we leverage the intrinsic potential that mode‐directed tabling has to express dynamic programming problems by creating a new design that improves the representation of multi‐dimensional arrays in the context of multithreaded tabling. To do so, we introduce a new mode for indexing arguments in mode‐directed tabled evaluations, named dim , where each dim argument features a uni‐dimensional lock‐free array. Experimental results using well‐known dynamic programming problems on a 32‐core machine show that the new design introduces less overheads and clearly improves the execution time for sequential and multithreaded tabled evaluations.

Areias, Miguel↗

Autotuning PolyBench benchmarks with LLVM Clang/Polly loop optimization pragmas using Bayesian optimization

Here, we develop a ytopt autotuning framework that leverages Bayesian optimization to explore the parameter space search and compare four different supervised learning methods within Bayesian optimization and evaluate their effectiveness. We select six of the most complex PolyBench benchmarks and apply the newly developed LLVM Clang/Polly loop optimization pragmas to the benchmarks to optimize them. We then use the autotuning framework to optimize the pragma parameters to improve their performance. The experimental results show that our autotuning approach outperforms the other compiling methods to provide the smallest execution time for the benchmarks syr2k, 3mm, heat-3d, lu, and covariance with two large datasets in 200 code evaluations for effectively searching the parameter spaces with up to 170,368 different configurations. We find that the Floyd-Warshall benchmark did not benefit from autotuning. To cope with this issue, we provide some compiler option solutions to improve the performance. Then we present loop autotuning without a user's knowledge using a simple mctree autotuning framework to further improve the performance of the Floyd-Warshall benchmark. We also extend the ytopt autotuning framework to tune a deep learning application.

79 ASTRONOMY AND ASTROPHYSICS↗

Results of a Geant4 benchmarking study for bio‐medical applications, performed with the G4‐Med system

Geant4, a Monte Carlo Simulation Toolkit extensively used in bio-medical physics, is in continuous evolution to include newest research findings to improve its accuracy and to respond to the evolving needs of a very diverse user community. In 2014, the G4-Med benchmarking system was born from the effort of the Geant4 Medical Simulation Benchmarking Group, to benchmark and monitor the evolution of Geant4 for medical physics applications. The G4-Med system was first described in our Medical Physics Special Report published in 2021. Results of the tests were reported for Geant4 10.5. Purpose In this work, we describe the evolution of the G4-Med benchmarking system. Methods The G4-Med benchmarking suite currently includes 23 tests, which benchmark Geant4 from the calculation of basic physical quantities to the simulation of more clinically relevant set-ups. New tests concern the benchmarking of Geant4-DNA physics and chemistry components for regression testing purposes, dosimetry for brachytherapy with a 125 I source, dosimetry for external x-ray and electron FLASH radiotherapy, experimental microdosimetry for proton therapy, and in vivo PET for carbon and oxygen beams. Regression testing has been performed between Geant4 10.5 and 11.1. Finally, a simple Geant4 simulation has been developed and used to compare Geant4 EM physics constructors and physics lists in terms of execution times. Results In summary, our EM tests show that the parameters of the multiple scattering in the Geant4 EM constructor G4EmStandardPhysics_option3 in Geant4 11.1, while improving the modeling of the electron backscattering in high atomic number targets, are not adequate for dosimetry for clinical x-ray and electron beams. Therefore, these parameters have been reverted back to those of Geant4 10.5 in Geant4 11.2.1. The x-ray radiotherapy test shows significant differences in the modeling of the bremsstrahlung process, especially between G4EmPenelopePhysics and the other constructors under study (G4EmLivermorePhysics, G4EmStandardPhysics_option3, and G4EmStandardPhysics_option4). These differences will be studied in an in-depth investigation within our Group. Improvement in Geant4 11.1 has been observed for the modeling of the proton and carbon ion Bragg peak with energies of clinical interest, thanks to the adoption of ICRU90 to calculate the low energy proton stopping powers in water and of the Linhard–Sorensen ion model, available in Geant4 since version 11.0. Nuclear fragmentation tests of interest for carbon ion therapy show differences between Geant4 10.5 and 11.1 in terms of fragment yields. In particular, a higher production of boron fragments is observed with Geant4 11.1, leading to a better agreement with reference data for this fragment. Conclusions Based on the overall results of our tests, we recommend to use G4EmStandardPhysics_option4 as EM constructor and QGSP_BIC_HP with G4EmStandardPhysics_option4, for hadrontherapy applications. The Geant4-DNA physics lists report differences in modeling electron interactions in water, however, the tests have a pure regression testing purpose so no recommendation can be formulated.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Pairwise‐Parallel Entangling Gates on Orthogonal Modes in a Trapped‐Ion Chain

Abstract Parallel operations are important for both near‐term quantum computers and larger‐scale fault‐tolerant machines because they reduce execution time and qubit idling. This study proposes and implements a pairwise‐parallel gate scheme on a trapped‐ion quantum computer. The gates are driven simultaneously on different sets of orthogonal motional modes of a trapped‐ion chain. This work demonstrates the utility of this scheme by creating a Greenberger‐Horne‐Zeilinger (GHZ) state in one step using parallel gates with one overlapping qubit. It also shows its advantage for circuits by implementing a digital quantum simulation of the dynamics of an interacting spin system, the transverse‐field Ising model. This method effectively extends the available gate depth by up to two times with no overhead when no overlapping qubit is involved, apart from additional initial cooling. This scheme can be easily applied to different trapped‐ion qubits and gate schemes, broadly enhancing the capabilities of trapped‐ion quantum computers.

Optics↗

Evaluating Performance Portability of Accelerator Programming Models using SPEC ACCEL 1.2 Benchmarks

As heterogeneous architectures are becoming mainstream for HPC systems, application programmers are looking for programming model implementations that offer both performance and portability across platforms. Two directive-based programming models for accelerator programming that aim at doing this are OpenMP 4/4.5 and OpenACC. Many users want to know the difference between these two programming models, the state of their implementations, how to use them, and evaluate how suitable they are for their applications.The Standard Performance Evaluation Corporation (SPEC) ACCEL benchmarks, developed by the SPEC High Performance Group (HPG), recently released SPEC ACCEL 1.2 benchmark suite to help the evaluation of OpenCL, OpenMP 4.5 and OpenACC on different platforms. Inthis paper we present our preliminary results that evaluates OpenMP4.5 and OpenACC on a variety of accelerator-based systems: POWER9with NVIDIA V100 GPUs (Summit), Intel Xeon Phi 7230 (Percival), and AMD Bulldozer Opteron with NVIDIA K20x (Titan). Comparing these benchmarks on different systems gives us insight into the support for OpenMP and OpenACC and their execution times provide insights about their quality of implementations provided by different vendors. We also compare best of OpenMP and OpenACC to see if a particular programming model favors a particular type of benchmark kernel.

Boehm, Swen↗

3D Coded SUMMA: Communication-Efficient and Robust Parallel Matrix Multiplication

In this paper, we propose a novel fault-tolerant parallel matrix multiplication algorithm called 3D Coded SUMMA that achieves higher failure-tolerance than replication-based schemes for the same amount of redundancy. This work bridges the gap between recent developments in coded computing and fault-tolerance in high-performance computing (HPC). The core idea of coded computing is the same as algorithm-based fault-tolerance (ABFT), which is weaving redundancy in the computation using error-correcting codes. In particular, we show that MatDot codes, an innovative code construction for parallel matrix multiplications, can be integrated into three-dimensional SUMMA (Scalable Universal Matrix Multiplication Algorithm [30]) in a communication-avoiding manner. To tolerate any two node failures, the proposed 3D Coded SUMMA requires ~50% less redundancy than replication, while the overhead in execution time is only about 5–10%.

97 MATHEMATICS AND COMPUTING↗