Search NASA⌕ Search

SEARCH · Search NASA

Results for “OpenMP parallelization”

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 127 records · Page 7

Effective Vectorization with OpenMP 4.5

This paper describes how the Single Instruction Multiple Data (SIMD) model and its extensions in OpenMP work, and how these are implemented in different compilers. Modern processors are highly parallel computational machines which often include multiple processors capable of executing several instructions in parallel. Understanding SIMD and executing instructions in parallel allows the processor to achieve higher performance without increasing the power required to run it. SIMD instructions can significantly reduce the runtime of code by executing a single operation on large groups of data. The SIMD model is so integral to the processor s potential performance that, if SIMD is not utilized, less than half of the processor is ever actually used. Unfortunately, using SIMD instructions is a challenge in higher level languages because most programming languages do not have a way to describe them. Most compilers are capable of vectorizing code by using the SIMD instructions, but there are many code features important for SIMD vectorization that the compiler cannot determine at compile time. OpenMP attempts to solve this by extending the C++/C and Fortran programming languages with compiler directives that express SIMD parallelism. OpenMP is used to pass hints to the compiler about the code to be executed in SIMD. This is a key resource for making optimized code, but it does not change whether or not the code can use SIMD operations. However, in many cases critical functions are limited by a poor understanding of how SIMD instructions are actually implemented, as SIMD can be implemented through vector instructions or simultaneous multi-threading (SMT). We have found that it is often the case that code cannot be vectorized, or is vectorized poorly, because the programmer does not have sufficient knowledge of how SIMD instructions work.

97 MATHEMATICS AND COMPUTING↗

Performance Characteristics of HYDRA - a Multi-Physics simulation code from Lawrence Livermore National Laboratory

HYDRA is used to simulate a variety of experiments carried out at the National Ignition Facility (NIF) [4] and other high energy density physics facilities. HYDRA has packages to simulate radiation transfer, atomic physics, hydrodynamics, laser propagation, and a number of other physics effects. HYDRA has over one million lines of code and includes both MPI and thread-level (OpenMP and pthreads) parallelism. This paper measures the performance characteristics of HYDRA using hardware counters on an IBM BlueGene/Q system. We report key ratios such as bytes/instruction and memory bandwidth for several different physics packages. The total number of bytes read and written per time step is also reported. We show that none of the packages which use significant time are memory bandwidth limited on a Blue Gene/Q. HYDRA currently issues very few SIMD instructions. The pressure on memory bandwidth will increase if high levels of SIMD instructions can be achieved.

97 MATHEMATICS AND COMPUTING↗

MODULES FOR EXPERIMENTS IN STELLAR ASTROPHYSICS (MESA)

Stellar physics and evolution calculations enable a broad range of research in astrophysics. Modules for Experiments in Stellar Astrophysics (MESA) is a suite of open source, robust, efficient, thread-safe libraries for a wide range of applications in computational stellar astrophysics. A one-dimensional stellar evolution module, MESAstar, combines many of the numerical and physics modules for simulations of a wide range of stellar evolution scenarios ranging from very low mass to massive stars, including advanced evolutionary phases. MESAstar solves the fully coupled structure and composition equations simultaneously. It uses adaptive mesh refinement and sophisticated timestep controls, and supports shared memory parallelism based on OpenMP. State-of-the-art modules provide equation of state, opacity, nuclear reaction rates, element diffusion data, and atmosphere boundary conditions. Each module is constructed as a separate Fortran 95 library with its own explicitly defined public interface to facilitate independent development. Several detailed examples indicate the extensive verification and testing that is continuously performed and demonstrate the wide range of capabilities that MESA possesses. These examples include evolutionary tracks of very low mass stars, brown dwarfs, and gas giant planets to very old ages; the complete evolutionary track of a 1 M {sub sun} star from the pre-main sequence (PMS) to a cooling white dwarf; the solar sound speed profile; the evolution of intermediate-mass stars through the He-core burning phase and thermal pulses on the He-shell burning asymptotic giant branch phase; the interior structure of slowly pulsating B Stars and Beta Cepheids; the complete evolutionary tracks of massive stars from the PMS to the onset of core collapse; mass transfer from stars undergoing Roche lobe overflow; and the evolution of helium accretion onto a neutron star. MESA can be downloaded from the project Web site (http://mesa.sourceforge.net/).

79 ASTRONOMY AND ASTROPHYSICS↗

Emu v1.1

Emu is a particle-in-cell code for solving the neutrino quantum kinetic equations in 1, 2, or 3 spatial dimensions with arbitrary angular resolution in order to simulate neutrino flavor transformation in neutron star merger and core collapse supernova environments. Emu represents the neutrino distribution function as a set of particles, each of which represent a collection of neutrinos and antineutrinos with unique position and momentum. Each particle carries two density matrices to define the flavor state of the neutrinos and antineutrinos it represents. Emu includes the vacuum, matter, and neutrino self-interaction potentials. Emu calculates the self-interaction potential using PIC deposition and interpolation algorithms that efficiently compute the local number density and flux at particle locations. Emu is implemented in C++ and is based on the AMReX library for high-performance, block-structured adaptive mesh refinement. Emu is parallelized with MPI + OpenMP for CPUs and MPI + CUDA for GPUs.

Willcox, Donald↗

R-Stream 3.22.3-DOE

R-stream release 3.22.3 including the Glow and TVM front-ends, and the features developed in the Portacy 2B, X-MAT and STAPL projects. R-Stream is an automatic parallelization, source-to-source compiler. The TVM and Glow front-ends to R-Stream were developed to automatically map neural net models to parallel computers using openMP and CUDA (for Glow).

Meister, Benoit↗

A Case Study of LLVM-Based Analysis for Optimizing SIMD Code Generation

This paper presents a methodology for using LLVM-based tools to tune the DCA++ (dynamical cluster approximation) application that targets the new ARM A64FX processor. The goal is to describe the changes required for the new architecture and generate efficient single instruction/multiple data (SIMD) instructions that target the new Scalable Vector Extension instruction set. During manual tuning, the authors used the LLVM tools to improve code parallelization by using OpenMP SIMD, refactored the code and applied transformation that enabled SIMD optimizations, and ensured that the correct libraries were used to achieve optimal performance. By applying these code changes, code speed was increased by 1.98× and 78 GFlops were achieved on the A64FX processor. The authors aim to automatize parts of the efforts in the OpenMP Advisor tool, which is built on top of existing and newly introduced LLVM tooling.

Huber, Joseph↗

ABWR start-up test analysis using BWR core simulator with three-dimensional direct response matrix method

The ABWR start-up test analysis has been done with the BWR core simulator using the three--dimensional direct response matrix (3D-DRM) method. The Monte Carlo code VMONT made the sub-response matrices for the 3D-DRM method. Each boundary surface was subdivided by 4 x 4 for transverse segments, by 4 for angular segments and by 4 for axial zones in a node. For the calculation speedup, the 3D-DRM code used the divided sub-response matrices data set. The code used the MPI and OpenMP for the parallelized method. The median value is set as the average critical eigenvalues. The changes from the maximum value to the minimum value are 0.34 %{Delta}k with the spectral history method and 0.40 %{Delta}k without it, and the respective standard deviations were 0.12 % and 0.14 %. Using the spectral history method decreased the variation by 0.06 %{Delta}k. The root mean square differences of the axial power distribution were about 6 % between the analysis results and the plant data. Using the currents which converged in the previous exposure step reduced the number of iterations when the CR pattern changed only slightly. The averaged calculation time for each exposure step was about 5 hours on 12 PC Linux cluster servers with Core 2 Quad 3 GHz. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

On the adequacy of lightweight thread approaches for high-level parallel programming models

High-level parallel programming models (PMs) are becoming crucial in order to extract the computational power of current on-node multi-threaded parallelism. The most popular PMs, such as OpenMP or OmpSs, are directive-based: the complexity of the hardware is hidden by the underlying runtime system, improving coding productivity. The implementations of OpenMP usually rely on POSIX threads (pthreads), offering excellent performance for coarse-grained parallelism and a perfect match with the current hardware. OmpSs is a task oriented PM based on an ad hoc runtime solution called Nanos++; it is the precursor of the tasking parallelism in the OpenMP tasking specification. A recent trend in runtimes and applications points to leveraging massive on-node parallelism in conjunction with fine-grained and dynamic scheduling paradigms. In this paper we analyze the behavior of the OpenMP and OmpSs PMs on top of the recently emerged Generic Lightweight Threads (GLT) API. GLT exposses a common API for lightweight thread (LWT) libraries that offers the possibility of running the same application over different native LWT solutions. We describe the design details of those high-level PMs implemented on top of GLT and analyze different scenarios in order to assess where the use of LWTs may benefit application performance. Our work reveals those scenarios where LWTs overperform pthread-based solutions and compares the performance between an ad hoc solution and a generic implementation.

97 MATHEMATICS AND COMPUTING↗

Automation of Data Traffic Control on DSM Architecture

The design of distributed shared memory (DSM) computers liberates users from the duty to distribute data across processors and allows for the incremental development of parallel programs using, for example, OpenMP or Java threads. DSM architecture greatly simplifies the development of parallel programs having good performance on a few processors. However, to achieve a good program scalability on DSM computers requires that the user understand data flow in the application and use various techniques to avoid data traffic congestions. In this paper we discuss a number of such techniques, including data blocking, data placement, data transposition and page size control and evaluate their efficiency on the NAS (NASA Advanced Supercomputing) Parallel Benchmarks. We also present a tool which automates the detection of constructs causing data congestions in Fortran array oriented codes and advises the user on code transformations for improving data traffic in the application.

Frumkin, Michael↗

Parallel peak pruning for scalable SMP contour tree computation

As data sets grow to exascale, automated data analysis and visualisation are increasingly important, to intermediate human understanding and to reduce demands on disk storage via in situ analysis. Trends in architecture of high performance computing systems necessitate analysis algorithms to make effective use of combinations of massively multicore and distributed systems. One of the principal analytic tools is the contour tree, which analyses relationships between contours to identify features of more than local importance. Unfortunately, the predominant algorithms for computing the contour tree are explicitly serial, and founded on serial metaphors, which has limited the scalability of this form of analysis. While there is some work on distributed contour tree computation, and separately on hybrid GPU-CPU computation, there is no efficient algorithm with strong formal guarantees on performance allied with fast practical performance. Here in this paper, we report the first shared SMP algorithm for fully parallel contour tree computation, withfor-mal guarantees of O(lgnlgt) parallel steps and O(n lgn) work, and implementations with up to 10x parallel speed up in OpenMP and up to 50x speed up in NVIDIA Thrust.

97 MATHEMATICS AND COMPUTING↗

Parallelizable adjoint stencil computations using transposed forward-mode algorithmic differentiation

Algorithmic differentiation (AD) is a tool for generating discrete adjoint solvers, which efficiently compute gradients of functions with many inputs, for example for use in gradient-based optimization. AD is often applied to large computations such as stencil operators, which are an important part of most structured-mesh PDE solvers. Stencil computations are often parallelized, for example by using OpenMP, and optimized by using techniques such as cache-blocking and tiling to fully utilize multicore CPUs and many-core accelerators and GPUs. Differentiating these codes with conventional reverse-mode AD results in adjoint codes that cannot be expressed as stencil operations and may not be easily parallelizable. They thus leave most of the compute power of modern architectures unused. We present a method that combines forward-mode AD and loop transformation to generate adjoint solvers that use the same memory access pattern as the original computation that they are derived from and can benefit from the same optimization techniques. The effectiveness of this method is demonstrated by generating a scalable adjoint CFD solver for multicore CPUs and Xeon Phi accelerators.

algorithmic differentiation↗

Analysis of Threading Libraries for High Performance Computing

With the appearance of multi-/many core machines, applications and runtime systems have evolved in order to exploit the new on-node concurrency brought by new software paradigms. POSIX threads (Pthreads) was widely-adopted for that purpose and it remains as the most used threading solution in current hardware. Lightweight thread (LWT) libraries emerged as an alternative offering lighter mechanisms to tackle the massive concurrency of current hardware. In this article, we analyze in detail the most representative threading libraries including Pthread- and LWT-based solutions. In addition, to examine the suitability of LWTs for different use cases, we develop a set of microbenchmarks consisting of OpenMP patterns commonly found in current parallel codes, and we compare the results using threading libraries and OpenMP implementations. Moreover, we study the semantics offered by threading libraries in order to expose the similarities among different LWT application programming interfaces and their advantages over Pthreads. This article exposes that LWT libraries outperform solutions based on operating system threads when tasks and nested parallelism are required.

GLT↗

A Conservative, Scalable, Space-Time Blade Element Rotor Model for Multi-Rotor Vehicles

The development of a parallel blade-element rotor model and its implementation into an adaptive Cartesian method is described. The unsteady version of the rotor model applies a body force to all cells contained in the swept space-time volume at each timestep and special care is taken to maintain axisymmetry on the Cartesian grid. Mesh convergence of rotor thrust and torque is obtained with around 10000 cells in the disk for the steady model. Parallelization is accomplished using OpenMP and the rotor force computation is distributed across all available nodes. Simulations of an isolated XV-15 rotor in hover show good correlation with experimental data and predictions of multi-rotor thrust variation closely match previous high fidelity simulations. The final paper will also include results from the unsteady rotor model and parallel scaling tests.

Conservative↗

A Massively Parallel Implementation of the CCSD(T) Method Using the Resolution-of-the-Identity Approximation and a Hybrid Distributed/Shared Memory Parallelization Model

In this work, a parallel algorithm is described for the coupled-cluster singles and doubles method augmented with a perturbative correction for triple excitations [CCSD(T)] using the resolution-of-the-identity (RI) approximation for two-electron repulsion integrals (ERIs). The algorithm bypasses the storage of four-center ERIs by adopting an integral-direct strategy. The CCSD amplitude equations are given in a compact quasi-linear form by factorizing them in terms of amplitude-dressed three-center intermediates. A hybrid MPI/OpenMP parallelization scheme is employed, which uses the OpenMP-based shared memory model for intranode parallelization and the MPI-based distributed memory model for internode parallelization. Parallel efficiency has been optimized for all terms in the CCSD amplitude equations. Two different algorithms have been implemented for the rate-limiting terms in the CCSD amplitude equations that entail and -scaling computational costs, where N O and N V denote the number of correlated occupied and virtual orbitals, respectively. One of the algorithms assembles the four-center ERIs requiring N V 4 and N O 2 N V 2 -scaling memory costs in a distributed manner on a number of MPI ranks, while the other algorithm completely bypasses the assembling of quartic memory-scaling ERIs and thus largely reduces the memory demand. It is demonstrated that the former memory-expensive algorithm is faster on a few hundred cores, while the latter memory-economic algorithm shows a better strong scaling in the limit of a few thousand cores. The program is shown to exhibit a near-linear scaling, in particular for the compute-intensive triples correction step, on up to 8000 cores. The performance of the program is demonstrated via calculations involving molecules with 24–51 atoms and up to 1624 atomic basis functions. As the first application, the complete basis set (CBS) limit for the interaction energy of the π-stacked uracil dimer from the S66 data set has been investigated. This work reports the first calculation of the interaction energy at the CCSD(T)/aug-cc-pVQZ level without local orbital approximation. The CBS limit for the CCSD correlation contribution to the interaction energy was found to be -8.01 kcal/mol, which agrees very well with the value -7.99 kcal/mol reported by Schmitz, Hättig, and Tew [ Phys. Chem. Chem. Phys. 2014 , 16 , 22167-22178]. The CBS limit for the total interaction energy was estimated to be -9.64 kcal/mol.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Hyperspectral Image Classification Using Graph Clustering Methods

Hyperspectral imagery is a challenging modality due to the dimension of the pixels which can range from hundreds to over a thousand frequencies depending on the sensor. Most methods in the literature reduce the dimension of the data using a method such as principal component analysis, however this procedure can lose information. More recently methods have been developed to address classification of large datasets in high dimensions. This paper presents two classes of graph-based classification methods for hyperspectral imagery. Using the full dimensionality of the data, we consider a similarity graph based on pairwise comparisons of pixels. The graph is segmented using a pseudospectral algorithm for graph clustering that requires information about the eigenfunctions of the graph Laplacian but does not require computation of the full graph. We develop a parallel version of the Nyström extension method to randomly sample the graph to construct a low rank approximation of the graph Laplacian. With at most a few hundred eigenfunctions, we can implement the clustering method designed to solve a variational problem for a graph-cut-based semi-supervised or unsupervised classification problem. We implement OpenMP directive-based parallelism in our algorithms and show performance improvement and strong, almost ideal, scaling behavior. The method can handle very large datasets including a video sequence with over a million pixels, and the problem of segmenting a data set into a pre-determined number of classes.

47 OTHER INSTRUMENTATION↗

Porting DMRG++ Scientific Application to OpenPOWER

With the rapidly changing microprocessor designs and architectural diversity (multi-cores, many-cores, accelerators) for the next generation HPC systems, scientific applications must adapt to the hardware, to exploit the different types of parallelism and resources available in the architecture. To get the benefit of all the in-node hardware threads, it is important to use a single programming model to map and coordinate the available work to the different heterogeneous execution units in the node (e.g., multi-core hardware threads (latency optimized), accelerators (bandwidth optimized), etc.).Our goal is to show that we can manage the node complexity of these systems by using OpenMP for in-node parallelization by exploiting different “programming styles” supported by OpenMP 4.5 to program CPU cores and accelerators. Finding out the suitable programming-style (e.g., SPMD style, multi-level tasks, accelerator programming, nested parallelism, or a combination of these) using the latest features of OpenMP to maximize performance and achieve performance portability across heterogeneous and homogeneous systems is still an open research problem.We developed a mini-application, Kronecker Product (KP), from the original DMRG++ application (sparse matrix algebra) computational motif to experiment with different OpenMP programming styles on an OpenPOWER architecture and present their results in this paper.

Chatterjee, Ronnie↗

The Ongoing Evolution of OpenMP

Here, this paper presents an overview of the past, present and future of the OpenMP application programming interface (API). While the API originally specified a small set of directives that guided shared memory fork-join parallelization of loops and program sections, OpenMP now provides a richer set of directives that capture a wide range of parallelization strategies that are not strictly limited to shared memory. As we look toward the future of OpenMP, we immediately see further evolution of the support for that range of parallelization strategies and the addition of direct support for debugging and performance analysis tools. Looking beyond the next major release of the specification of the OpenMP API, we expect the specification eventually to include support for more parallelization strategies and to embrace closer integration into its Fortran, C and, in particular, C++ base languages, which will likely require the API to adopt additional programming abstractions.

97 MATHEMATICS AND COMPUTING↗