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 343 records · Page 19

HipMCL: a high-performance parallel implementation of the Markov clustering algorithm for large-scale networks

Biological networks capture structural or functional properties of relevant entities such as molecules, proteins or genes. Characteristic examples are gene expression networks or protein–protein interaction networks, which hold information about functional affinities or structural similarities. Such networks have been expanding in size due to increasing scale and abundance of biological data. While various clustering algorithms have been proposed to find highly connected regions, Markov Clustering (MCL) has been one of the most successful approaches to cluster sequence similarity or expression networks. Despite its popularity, MCL’s scalability to cluster large datasets still remains a bottleneck due to high running times and memory demands. In this paper, we present High-performance MCL (HipMCL), a parallel implementation of the original MCL algorithm that can run on distributed-memory computers. We show that HipMCL can efficiently utilize 2000 compute nodes and cluster a network of ~70 million nodes with ~68 billion edges in ~2.4 h. By exploiting distributed-memory environments, HipMCL clusters large-scale networks several orders of magnitude faster than MCL and enables clustering of even bigger networks. Finally, HipMCL is based on MPI and OpenMP and is freely available under a modified BSD license.

59 BASIC BIOLOGICAL SCIENCES↗

A communication-avoiding 3D sparse triangular solver

We present a novel distributed memory algorithm to improve the strong scalability of the solution of a sparse triangular system. This operation appears in the solve phase of direct methods for solving general sparse linear systems, Ax = b. Our 3D sparse triangular solver employs several techniques, including a 3D MPI process grid, elimination tree parallelism, and data replication, all of which reduce the per-process communication when combined. We present analytical models to understand the communication cost of our algorithm and show that our 3D sparse triangular solver can reduce the per-process communication volume asymptotically by a factor of O(n1/4) and O(n1/6) for problems arising from the finite element discretizations of 2D "planar" and 3D "non-planar" PDEs, respectively. We implement our algorithm for use in SuperLU_DIST3D, using a hybrid MPI+OpenMP programming model. Our 3D triangular solve algorithm, when run on 12k cores of Cray XC30, outperforms the current state-of-the-art 2D algorithm by 7.2x for planar and 2.7x for the non-planar sparse matrices, respectively.

Sao, Piyush↗

Hybrid programming-model strategies for GPU offloading of electronic structure calculation kernels

To address the challenge of performance portability and facilitate the implementation of electronic structure solvers, we developed the basic matrix library (BML) and Parallel, Rapid O(N), and Graph-based Recursive Electronic Structure Solver (PROGRESS) library. The BML implements linear algebra operations necessary for electronic structure kernels using a unified user interface for various matrix formats (dense and sparse) and architectures (CPUs and GPUs). Focusing on density functional theory and tight-binding models, PROGRESS implements several solvers for computing the single-particle density matrix and relies on BML. In this paper, we describe the general strategies used for these implementations on various computer architectures, using OpenMP target functionalities on GPUs, in conjunction with third-party libraries to handle performance critical numerical kernels. In this study, we demonstrate the portability of this approach and its performance in benchmark problems.

36 MATERIALS SCIENCE↗

GronOR: Massively Parallel and GPU-Accelerated Non-Orthogonal Configuration Interaction for Large Molecular Systems

GronOR is a program package for non-orthogonal configuration interaction calculations for an electronic wave function built in terms of anti-symmetrized products of multi-configuration molecular fragment wave functions. The two-electron integrals that have to be processed may be expressed in terms of atomic orbitals or in terms of an orbital basis determined from the molecular orbitals of the fragments. The code has been specifically designed for execution on distributed memory massively parallel and Graphics Processing Unit (GPU)-accelerated computer architectures, using an MPI+OpenACC/OpenMP programming approach. The task-based execution model used in the implementation allows for linear scaling with the number of nodes on the largest pre-exascale architectures available, provides hardware fault resiliency, and enables effective execution on systems with distinct central processing unit-only and GPU-accelerated partitions. The code interfaces with existing multi-configuration electronic structure codes that provide optimized molecular fragment orbitals, configuration interaction coefficients, and the required integrals. Algorithm and implementation details, parallel and accelerated performance benchmarks, and an analysis of the sensitivity of the accuracy of results and computational performance to thresholds used in the calculations are presented.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Gregarious Data Re-structuring in a Many Core Architecture

this paper, we have developed a new methodology that takes in consideration the access patterns from a single parallel actor (e.g. a thread), as well as, the access patterns of “grouped” parallel actors that share a resource (e.g. a distributed Level 3 cache). We start with a hierarchical tile code for our target machine and apply a series of transformations at the tile level to improve data residence in a given memory hierarchy level. The contribution of this paper includes (a) collaborative data restructuring for group reuse and (b) low overhead transformation technique to improve access pattern and bring closely connected data elements together. Preliminary results in a many core architecture, Tilera TileGX, shows promising improvements over optimized OpenMP code (up to 31% increase in GFLOPS) and over our own previous work on fine grained runtimes (up to 16%) for selected kernels

data restructuring↗

A Comparative Critical Analysis of Modern Task-Parallel Runtimes

The rise in node-level parallelism has increased interest in task-based parallel runtimes for a wide array of application areas. Applications have a wide variety of task spawning patterns which frequently change during the course of application execution, based on the algorithm or solver kernel in use. Task scheduling and load balance regimes, however, are often highly optimized for specific patterns. This paper uses four basic task spawning patterns to quantify the impact of specific scheduling policy decisions on execution time. We compare the behavior of six publicly available tasking runtimes: Intel Cilk, Intel Threading Building Blocks (TBB), Intel OpenMP, GCC OpenMP, Qthreads, and High Performance ParalleX (HPX). With the exception of Qthreads, the runtimes prove to have schedulers that are highly sensitive to application structure. No runtime is able to provide the best performance in all cases, and those that do provide the best performance in some cases, unfortunately, provide extremely poor performance when application structure does not match the scheduler's assumptions.

Wheeler, Kyle Bruce↗

Power Analysis for Applications and Data Centers

A portion of a chapter for the High Performance Parallelism Pearls 2 book. By using the NWPerf and CView software developed and deployed at PNNL administrators and users can visually see the performance, and power used by jobs running on a super computer in near-time.

Affinity↗

Automatic Differentiation of C++ Codes on Emerging Manycore Architectures with Sacado

Automatic differentiation (AD) is a well-known technique for evaluating analytic derivatives of calculations implemented on a computer, with numerous software tools available for incorporating AD technology into complex applications. However, a growing challenge for AD is the efficient differentiation of parallel computations implemented on emerging manycore computing architectures such as multicore CPUs, GPUs, and accelerators as these devices become more pervasive. In this work, we explore forward mode, operator overloading-based differentiation of C++ codes on these architectures using the widely available Sacado AD software package. In particular, we leverage Kokkos, a C++ tool providing APIs for implementing parallel computations that is portable to a wide variety of emerging architectures. Here we describe the challenges that arise when differentiating code for these architectures using Kokkos, and two approaches for overcoming them that ensure optimal memory access patterns as well as expose additional dimensions of fine-grained parallelism in the derivative calculation. We describe the results of several computational experiments that demonstrate the performance of the approach on a few contemporary CPU and GPU architectures. We then conclude with applications of these techniques to the simulation of discretized systems of partial differential equations.

97 MATHEMATICS AND COMPUTING↗

KokkACC: Enhancing Kokkos with OpenACC

Template metaprogramming is gaining popularity as a high-level solution for achieving performance portability on heterogeneous computing resources. Kokkos is a representative approach that offers programmers high-level abstractions for generic programming while most of the device-specific code generation and optimizations are delegated to the compiler through template specializations. For this, Kokkos provides a set of device-specific code specializations in multiple back ends, such as CUDA and HIP. Unlike CUDA or HIP, OpenACC is a high-level and directive-based programming model. This descriptive model allows developers to insert hints (pragmas) into their code that help the compiler to parallelize the code. The compiler is responsible for the transformation of the code, which is completely transparent to the programmer. This paper presents an OpenACC back end for Kokkos: KokkACC. As an alternative to Kokkos’s existing device-specific back ends, KokkACC is a multi-architecture back end providing a high-productivity programming environment enabled by OpenACC’s high-level and descriptive programming model. Moreover, we have observed competitive performance; in some cases, KokkACC is faster (up to 9×) than NVIDIA’s CUDA back end and much faster than OpenMP’s GPU offloading back end. This work also includes implementation details and a detailed performance study conducted with a set of mini-benchmarks (AXPY and DOT product) and three mini-apps (LULESH, miniFE and SNAP, a LAMMPS proxy mini-app).

Valero Lara, Pedro↗

Getting To Exascale: Applying Novel Parallel Programming Models To Lab Applications For The Next Generation Of Supercomputers

As supercomputing moves toward exascale, node architectures will change significantly. CPU core counts on nodes will increase by an order of magnitude or more. Heterogeneous architectures will become more commonplace, with GPUs or FPGAs providing additional computational power. Novel programming models may make better use of on-node parallelism in these new architectures than do current models. In this paper we examine several of these novel models – UPC, CUDA, and OpenCL –to determine their suitability to LLNL scientific application codes. Our study consisted of several phases: We conducted interviews with code teams and selected two codes to port; We learned how to program in the new models and ported the codes; We debugged and tuned the ported applications; We measured results, and documented our findings. We conclude that UPC is a challenge for porting code, Berkeley UPC is not very robust, and UPC is not suitable as a general alternative to OpenMP for a number of reasons. CUDA is well supported and robust but is a proprietary NVIDIA standard, while OpenCL is an open standard. Both are well suited to a specific set of application problems that can be run on GPUs, but some problems are not suited to GPUs. Further study of the landscape of novel models is recommended.

97 MATHEMATICS AND COMPUTING↗

Portability for GPU-accelerated molecular docking applications for cloud and HPC: can portable compiler directives provide performance across all platforms?

High-throughput structure-based screening of drug-like molecules has become a common tool in biomedical research. Recently, acceleration with graphics processing units (GPUs) has provided a large performance boost for molecular docking programs. Both cloud and high-performance computing (HPC) resources have been used for large screens with molecular docking programs; while NVIDIA GPUs have dominated cloud and HPC resources, new vendors such as AMD and Intel are now entering the field, creating the problem of software portability across different GPUs. Ideally, software productivity could be maximized with portable programming models that are able to maintain high performance across architectures. While in many cases compiler directives have been used as an easy way to offload parallel regions of a CPU-based program to a GPU accelerator, they may also be an attractive programming model for providing portability across different GPU vendors, in which case the porting process may proceed in the reverse direction: from low-level, architecture-specific code to higher-level directive-based abstractions. MiniMDock is a new mini-application (miniapp) designed to capture the essential computational kernels found in molecular docking calculations, such as are used in phar-maceutical drug discovery efforts, in order to test different solutions for porting across GPU architectures. Here we extend MiniMDock to GPU offloading with OpenMP directives, and compare to performance of kernels using CUDA and HIP on NVIDIA and AMD GPUs, respectively, as well as across different compilers, exploring performance bottlenecks. We document this reverse-porting process, from highly optimized device code to a higher-level version using directives, compare code structure, and describe barriers that were overcome in this effort.

Thavappiragasam, Mathialakan↗

Accelerating Subsurface Transport Simulation on Heterogeneous Clusters

Reactive transport numerical models simulate chemical and microbiological reactions that occur along a flowpath. These models have to compute reactions for a large number of locations. They solve the set of ordinary differential equations (ODEs) that describes the reaction for each location through the Newton-Raphson technique. This technique involves computing a Jacobian matrix and a residual vector for each set of equation, and then solving iteratively the linearized system by performing Gaussian Elimination and LU decomposition until convergence. STOMP, a well known subsurface flow simulation tool, employs matrices with sizes in the order of 100x100 elements and, for numerical accuracy, LU factorization with full pivoting instead of the faster partial pivoting. Modern high performance computing systems are heterogeneous machines whose nodes integrate both CPUs and GPUs, exposing unprecedented amounts of parallelism. To exploit all their computational power, applications must use both the types of processing elements. For the case of subsurface flow simulation, this mainly requires implementing efficient batched LU-based solvers and identifying efficient solutions for enabling load balancing among the different processors of the system. In this paper we discuss two approaches that allows scaling STOMP's performance on heterogeneous clusters. We initially identify the challenges in implementing batched LU-based solvers for small matrices on GPUs, and propose an implementation that fulfills STOMP's requirements. We compare this implementation to other existing solutions. Then, we combine the batched GPU solver with an OpenMP-based CPU solver, and present an adaptive load balancer that dynamically distributes the linear systems to solve between the two components inside a node. We show how these approaches, integrated into the full application, provide speed ups from 6 to 7 times on large problems, executed on up to 16 nodes of a cluster with two AMD Opteron 6272 and a Tesla M2090 per node.

GPU↗

A Framework for the Analysis of Compiler Optimizations

Compilers transform program source code to machine executable code. During this transformation, they perform a number of compiler optimizations to improve the performance of the generated executable code. Importantly, applying those optimizations depends on the source code structure, such as the parallel programming model used to parallelize an algorithm. Often, implementations of the same algorithm with different programming models have vastly different performance because the compiler optimized them differently. We create FAROS, a framework to structure and automate the analysis of compiler optimizations on programs. FAROS automates the building process, execution profiling, and analysis of compiler optimization of programs, through a configuration interface. It outputs compiler optimization reports to show which optimizations applied to which line of source code, leveraging compilation remarks output by the compiler. Also, FAROS supports benchmarking performance of different program versions by collecting execution time results. In this first release of FAROS, we provide a configuration file to analyze compiler optimization differences for sequential vs OpenMP compilation, including 38 programs consisting of HPC proxy/mini/large applications, and NAS and Rodina kernels for analysis.

Georgakoudis, Giorgis↗

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↗

High Resolution Aerospace Applications using the NASA Columbia Supercomputer

This paper focuses on the parallel performance of two high-performance aerodynamic simulation packages on the newly installed NASA Columbia supercomputer. These packages include both a high-fidelity, unstructured, Reynolds-averaged Navier-Stokes solver, and a fully-automated inviscid flow package for cut-cell Cartesian grids. The complementary combination of these two simulation codes enables high-fidelity characterization of aerospace vehicle design performance over the entire flight envelope through extensive parametric analysis and detailed simulation of critical regions of the flight envelope. Both packages. are industrial-level codes designed for complex geometry and incorpor.ats. CuStomized multigrid solution algorithms. The performance of these codes on Columbia is examined using both MPI and OpenMP and using both the NUMAlink and InfiniBand interconnect fabrics. Numerical results demonstrate good scalability on up to 2016 CPUs using the NUMAIink4 interconnect, with measured computational rates in the vicinity of 3 TFLOP/s, while InfiniBand showed some performance degradation at high CPU counts, particularly with multigrid. Nonetheless, the results are encouraging enough to indicate that larger test cases using combined MPI/OpenMP communication should scale well on even more processors.

Mavriplis, Dimitri J.↗

Parallel k-means++ for Multiple Shared-Memory Architectures

In recent years k-means++ has become a popular initialization technique for improved k-means clustering. To date, most of the work done to improve its performance has involved parallelizing algorithms that are only approximations of k-means++. In this paper we present a parallelization of the exact k-means++ algorithm, with a proof of its correctness. We develop implementations for three distinct shared-memory architectures: multicore CPU, high performance GPU, and the massively multithreaded Cray XMT platform. We demonstrate the scalability of the algorithm on each platform. In addition we present a visual approach for showing which platform performed k-means++ the fastest for varying data sizes.

data clustering↗

Optimizing the hypre solver for manycore and GPU architectures

The solution of large-scale combustion problems with codes such as Uintah on modern computer architectures requires the use of multithreading and GPUs to achieve performance. Uintah uses a low-Mach number approximation that requires iteratively solving a large system of linear equations. The Hypre iterative solver has solved such systems in a scalable way for Uintah, but the use of OpenMP with Hypre leads to at least slowdown due to OpenMP overheads. The proposed solution uses the MPI Endpoints within Hypre, where each team of threads acts as a different MPI rank. This approach minimizes OpenMP synchronization overhead and performs as fast or (up to 1.44) faster than Hypre's MPI-only version, and allows the rest of Uintah to be optimized using OpenMP. The profiling of the GPU version of Hypre shows the bottleneck to be the launch overhead of thousands of micro-kernels. The GPU performance was improved by fusing these micro-kernels and was further optimized by using Cuda-aware MPI, resulting in an overall speedup of 1.16—1.44 compared to the baseline GPU implementation. The above optimization strategies were published in the International Conference on Computational Science 2020 [1]. This work extends the previously published research by carrying out the second phase of communication-centered optimizations in Hypre to improve its scalability on large-scale supercomputers. Additionally, this includes an efficient non-blocking inter-thread communication scheme, communication-reducing patch assignment, and expression of logical communication parallelism to a new version of the MPICH library that utilizes the underlying network parallelism [2]. The above optimizations avoid communication bottlenecks previously observed during strong scaling and improve performance by up to 2 on 256 nodes of Intel Knight's Landing processor.

97 MATHEMATICS AND COMPUTING↗

A Single Thread to Fortran Coarray Transition Process for the Control Algorithm in the Space Radiation Code HZETRN

Exa-scale computing is the direction by industry and government are going to generate solutions to problems they deem necessary. Computing hardware is being developed to achieve the transition from Peta-scale to Exa-scale with more CPUs (Central Processing Units) that have more cores per CPU and more accelerators (GPGPUs (General Purpose Graphics Processing Units) and MICs (Many Integrated Cores)) per node. To fully utilize the hardware available now and in the future, algorithms must become multi-threaded. There are a few methods to generate multi-threaded software such as MPI (Message Passing Interface) and OpenMP (Multi-Processing) / OpenACC (ACCelerator). This paper concentrates on using Coarray Fortran to convert the Fortran 95 based HZETRN (High Z and Energy TRaNsport) code's control algorithm from a single threaded code to a multithreaded code. The resultant Coarray code was 32.5 times faster (with a theoretical speed-up of 74.5 times) than the single threaded version on the hardware tested, as reliable as the Fortran 95 version, and, as it uses native Fortran, was as maintainable as the Fortran 95 version. The Coarray code can be maintained by the same project engineers and scientists who created the original single threaded code. This transition process can be utilized on a C language based code with a compiler that has the UPC (Universal Parallel C) extensions to C.

Singleterry, Robert C., Jr.↗