Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “message passing interface”

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 73 records · Page 4

Optimizing blocking and nonblocking reduction operations for multicore systems: Hierarchical design and implementation

Many scientific simulations, using the Message Passing Interface (MPI) programming model, are sensitive to the performance and scalability of reduction collective operations such as MPI_Allreduce and MPI_Reduce. These operations are the most widely used abstractions to perform mathematical operations over all processes that are part of the simulation. In this work, we propose a hierarchical design to implement the reduction operations on multicore systems. This design aims to improve the efficiency of reductions by 1) tailoring the algorithms and customizing the implementations for various communication mechanisms in the system 2) providing the ability to configure the depth of hierarchy to match the system architecture, and 3) providing the ability to independently progress each of this hierarchy. Using this design, we implement MPI_Allreduce and MPI_Reduce operations (and its nonblocking variants MPI_Iallreduce and MPI_Ireduce) for all message sizes, and evaluate on multiple architectures including InfiniBand and Cray XT5. We leverage and enhance our existing infrastructure, Cheetah, which is a framework for implementing hierarchical collective operations to implement these reductions. The experimental results show that the Cheetah reduction operations outperform the production-grade MPI implementations such as Open MPI default, Cray MPI, and MVAPICH2, demonstrating its efficiency, flexibility and portability. On Infini-Band systems, with a microbenchmark, a 512-process Cheetah nonblocking Allreduce and Reduce achieves a speedup of 23x and 10x, respectively, compared to the default Open MPI reductions. The blocking variants of the reduction operations also show similar performance benefits. A 512-process nonblocking Cheetah Allreduce achieves a speedup of 3x, compared to the default MVAPICH2 Allreduce implementation. On a Cray XT5 system, a 6144-process Cheetah Allreduce outperforms the Cray MPI by 145%. The evaluation with an application kernel, Conjugate Gradient solver, shows that the Cheetah reductions speeds up total time to solution by 195%, demonstrating the potential benefits for scientific simulations.

Venkata, Manjunath Gorentla↗

Optimizing Blocking and Nonblocking Reduction Operations for Multicore Systems: Hierarchical Design and Implementation

Many scientific simulations, using the Message Passing Interface (MPI) programming model, are sensitive to the performance and scalability of reduction collective operations such as MPI Allreduce and MPI Reduce. These operations are the most widely used abstractions to perform mathematical operations over all processes that are part of the simulation. In this work, we propose a hierarchical design to implement the reduction operations on multicore systems. This design aims to improve the efficiency of reductions by 1) tailoring the algorithms and customizing the implementations for various communication mechanisms in the system 2) providing the ability to configure the depth of hierarchy to match the system architecture, and 3) providing the ability to independently progress each of this hierarchy. Using this design, we implement MPI Allreduce and MPI Reduce operations (and its nonblocking variants MPI Iallreduce and MPI Ireduce) for all message sizes, and evaluate on multiple architectures including InfiniBand and Cray XT5. We leverage and enhance our existing infrastructure, Cheetah, which is a framework for implementing hierarchical collective operations to implement these reductions. The experimental results show that the Cheetah reduction operations outperform the production-grade MPI implementations such as Open MPI default, Cray MPI, and MVAPICH2, demonstrating its efficiency, flexibility and portability. On Infini- Band systems, with a microbenchmark, a 512-process Cheetah nonblocking Allreduce and Reduce achieves a speedup of 23x and 10x, respectively, compared to the default Open MPI reductions. The blocking variants of the reduction operations also show similar performance benefits. A 512-process nonblocking Cheetah Allreduce achieves a speedup of 3x, compared to the default MVAPICH2 Allreduce implementation. On a Cray XT5 system, a 6144-process Cheetah Allreduce outperforms the Cray MPI by 145%. The evaluation with an application kernel, Conjugate Gradient solver, shows that the Cheetah reductions speeds up total time to solution by 195%, demonstrating the potential benefits for scientific simulations.

Gorentla Venkata, Manjunath↗

Clock Synchronization in High-end Computing Environments: A Strategy for Minimizing Clock Variance at Runtime

We present a new software-based clock synchronization scheme that provides high precision time agreement among distributed memory nodes. The technique is designed to minimize variance from a reference chimer during runtime and with minimal time-request latency. Our scheme permits initial unbounded variations in time and corrects both slow and fast chimers (clock skew). An implementation developed within the context of the MPI message passing interface is described, and time coordination measurements are presented. Among our results, the mean time variance for a set of nodes improved from 20.0 milliseconds under standard Network Time Protocol (NTP) down to 2.29 secs under our scheme.

clock synchronization↗

Performance analysis of distributed symmetric sparse matrix vector multiplication algorithm for multi-core architectures

In this article, sparse matrix vector multiply (SpMVM) is an important kernel that frequently arises in high performance computing applications. Due to its low arithmetic intensity, several approaches have been proposed in literature to improve its scalability and efficiency in large scale computations. In this paper, our target systems are high end multi-core architectures and we use messaging passing interface + open multiprocessing hybrid programming model for parallelism. We analyze the performance of recently proposed implementation of the distributed symmetric SpMVM, originally developed for large sparse symmetric matrices arising in ab initio nuclear structure calculations. We also study important features of this implementation and compare with previously reported implementations that do not exploit underlying symmetry. Our SpMVM implementations leverage the hybrid paradigm to efficiently overlap expensive communications with computations. Our main comparison criterion is the "CPU core hours" metric, which is the main measure of resource usage on supercomputers. We analyze the effects of topology-aware mapping heuristic using simplified network load model. Furthermore, we have tested the different SpMVM implementations on two large clusters with 3D Torus and Dragonfly topology. Our results show that the distributed SpMVM implementation that exploits matrix symmetry and hides communication yields the best value for the "CPU core hours" metric and significantly reduces data movement overheads.

97 MATHEMATICS AND COMPUTING↗

A survey of MPI usage in the US exascale computing project

The Exascale Computing Project (ECP) is currently the primary effort in the United States focused on developing “exascale” levels of computing capabilities, including hardware, software, and applications. In order to obtain a more thorough understanding of how the software projects under the ECP are using, and planning to use the Message Passing Interface (MPI), and help guide the work of our own project within the ECP, we created a survey. Of the 97 ECP projects active at the time the survey was distributed, we received 77 responses, 56 of which reported that their projects were using MPI. Furthermore, this paper reports the results of that survey for the benefit of the broader community of MPI developers.

97 MATHEMATICS AND COMPUTING↗

On the memory attribution problem: A solution and case study using MPI

As parallel applications running on large–scale computing systems become increasingly memory constrained, the ability to attribute memory usage to the various components of the application is becoming increasingly important. We present the design and implementation of memnesia, a novel memory usage profiler for parallel and distributed message–passing applications. Our approach captures both application– and message–passing library–specific memory usage statistics from unmodified binaries dynamically linked to a message–passing communication library. Using microbenchmarks and proxy applications, we evaluated our profiler across three Message Passing Interface (MPI) implementations and two hardware platforms. Furthermore, the results show that our approach and the corresponding implementation can accurately quantify memory resource usage as a function of time, scale, communication workload, and software or hardware system architecture, clearly distinguishing between application and MPI library memory usage at a per–process level. With this new capability, we show that job size, communication workload, and hardware/software architecture influence peak runtime memory usage. In practice, this tool provides a potentially valuable source of information for application developers seeking to measure and optimize memory usage.

97 MATHEMATICS AND COMPUTING↗

Evaluating integration and performance of containerized climate applications on a Hewlett Packard Enterprise Cray system

Containers have taken over large swaths of cloud computing as the most convenient way of packaging and deploying applications. The features that containers offer for packaging and deploying applications translate to high performance computing (HPC) as well. At The National Oceanic and Atmospheric Administration, containers provide an easy way to build and distribute complex HPC applications, allowing faster collaboration, portability, and experiment computer environment reproducibility amongst the scientific community. The challenge arises when applications rely on message passing interface (MPI). This necessitates investigation into how to properly run these applications with their own unique requirements and produce performance on par with native runs. We investigate the MPI performance for benchmarks and containerized climate models for various containers covering selection of compiler and MPI library combinations from the Cray provided programming environments on the Cray XC supercomputer GAEA. Performance from the benchmarks and the climate models shows that for the most part containerized applications perform on par with the natively built applications when the system optimized Cray MPICH libraries are bound into the container, and the hybrid model containers have poor performance in comparison. We also describe several challenges and our solutions in running these containers, particularly challenges with heterogeneous jobs for the containerized model runs.

Abraham, Subil↗

OpenACC acceleration of an unstructured CFD solver based on a reconstructed discontinuous Galerkin method for compressible flows

Here, an OpenACC directive-based graphics processing unit (GPU) parallel scheme is presented for solving the compressible Navier–Stokes equations on 3D hybrid unstructured grids with a third-order reconstructed discontinuous Galerkin method. The developed scheme requires the minimum code intrusion and algorithm alteration for upgrading a legacy solver with the GPU computing capability at very little extra effort in programming, which leads to a unified and portable code development strategy. A face coloring algorithm is adopted to eliminate the memory contention because of the threading of internal and boundary face integrals. A number of flow problems are presented to verify the implementation of the developed scheme. Timing measurements were obtained by running the resulting GPU code on one Nvidia Tesla K20c GPU card (Nvidia Corporation, Santa Clara, CA, USA) and compared with those obtained by running the equivalent Message Passing Interface (MPI) parallel CPU code on a compute node (consisting of two AMD Opteron 6128 eight-core CPUs (Advanced Micro Devices, Inc., Sunnyvale, CA, USA)). Speedup factors of up to 24× and 1.6× for the GPU code were achieved with respect to one and 16 CPU cores, respectively. The numerical results indicate that this OpenACC-based parallel scheme is an effective and extensible approach to port unstructured high-order CFD solvers to GPU computing.

97 MATHEMATICS AND COMPUTING↗

VinaMPI: Facilitating multiple receptor high-throughput virtual docking on high-performance computers

The program VinaMPI has been developed to enable massively large virtual drug screens on leadership-class computing resources, using a large number of cores to decrease the time-to-completion of the screen. VinaMPI is a massively parallel Message Passing Interface (MPI) program based on the multithreaded virtual docking program AutodockVina, and is used to distribute tasks while multithreading is used to speed-up individual docking tasks. VinaMPI uses a distribution scheme in which tasks are evenly distributed to the workers based on the complexity of each task, as defined by the number of rotatable bonds in each chemical compound investigated. VinaMPI efficiently handles multiple proteins in a ligand screen, allowing for high-throughput inverse docking that presents new opportunities for improving the efficiency of the drug discovery pipeline. VinaMPI successfully ran on 84,672 cores with a continual decrease in job completion time with increasing core count. The ratio of the number of tasks in a screening to the number of workers should be at least around 100 in order to have a good load balance and an optimal job completion time. The code is freely available and downloadable. Lastly, the instructions for downloading and using the code are provided in the Supporting Information.

97 MATHEMATICS AND COMPUTING↗

A Parallel Computing Framework for Finding the Supported Solutions to a Biobjective Network Optimization Problem

Abstract Solving multi‐objective combinatorial optimization problems can quickly become computationally challenging when applied to large networks generated from big data. Present‐day first‐rate Mixed Integer Programming (MIP) solvers have parallelism built‐in to take advantage of multicore architectures; but specialized network optimization algorithms that can often solve graph problems more efficiently than a general MIP solver are typically programmed serially. Thus, these network algorithm implementations do not take full advantage of modern multicore computer capabilities. Many parallel computing languages include a fork/join framework into their concurrency packages that allow for simple conversion of recursive serial algorithms to operate in parallel without having to use complicated low‐level message‐passing interfaces. Fork/join is particularly well suited to divide and conquer algorithms such as Non‐Inferior Set Estimation (NISE) method for computing the supported Pareto front of a multi‐objective optimization problem. This work develops a simple general parallel NISE (pNISE) implementation using the Java 7 concurrency tools, and then offers results from a specific implementation of solving a bi‐objective shortest path problem. The results indicate that this method is capable of sizeable speed‐ups on both small‐scale personal computers as well as large‐scale shared memory supercomputers. Finally, other network problems suitable to this method are discussed, including multi‐objective variants of the minimum spanning tree problem, transportation problem, assignment problem, maximum flow problem, and the minimum‐cost flow problem. Copyright © 2015 John Wiley & Sons, Ltd.

Medrano, Fernando Antonio↗

OpenMP in VASP: Threading and SIMD

Abstract The Vienna Ab initio Simulation Package (VASP) is a widely used electronic structure code that originally exploits process‐level parallelism through the Message Passing Interface (MPI) for work distribution within and across nodes. Architectural changes of modern parallel processors urge programmers to address thread‐ and data‐level parallelism as well to benefit most from the available compute resources within a node. We describe for VASP how to approach for an MPI + OpenMP parallelization including data‐level parallelism through OpenMP SIMD constructs together with a generic high‐level vector coding scheme. We can demonstrate an improved scalability of VASP and more than 20% gain over the MPI‐only version as well as a 2× increased performance of collective operations using the multiple‐endpoint MPI feature. The high‐level vector coding scheme applied to VASP's general gradient approximation routine gives up 9× performance gain on AVX512 platforms with the Intel compiler.

Wende, Florian↗

VeloxChem: A Python-driven density-functional theory program for spectroscopy simulations in high-performance computing environments

An open-source program named VeloxChem has been developed for the calculation of electronic real and complex linear response functions at the levels of Hartree–Fock and Kohn–Sham density functional theories. With an object-oriented program structure written in a Python/C++ layered fashion, VeloxChem enables time-efficient prototyping of novel scientific approaches without sacrificing computational efficiency, so that molecular systems involving up to and beyond 500 second-row atoms (or some 10,000 contracted and in part diffuse Gaussian basis functions) can be routinely addressed. In addition, VeloxChem is equipped with a polarizable embedding scheme for the treatment of the classical electrostatic interactions with an environment that in turn is modeled by atomic site charges and polarizabilities. The underlying hybrid message passing interface (MPI)/open multiprocessing (OpenMP) parallelization scheme makes VeloxChem suitable for execution in high-performance computing cluster environments, showing even slightly beyond linear scaling for the Fock matrix construction with use of up to 16,384 central processing unit (CPU) cores. An efficient—with respect to convergence rate and overall computational cost—multifrequency/gradient complex linear response equation solver enables calculations not only of conventional spectra, such as visible/ultraviolet/X-ray electronic absorption and circular dichroism spectra, but also time-resolved linear response signals as due to ultra-short weak laser pulses.

97 MATHEMATICS AND COMPUTING↗

Roofline model toolkit: A practical tool for architectural and program analysis

We present preliminary results of the Roofline Toolkit for multicore, many core, and accelerated architectures. This paper focuses on the processor architecture characterization engine, a collection of portable instrumented micro benchmarks implemented with Message Passing Interface (MPI), and OpenMP used to express thread-level parallelism. These benchmarks are specialized to quantify the behavior of different architectural features. Compared to previous work on performance characterization, these microbenchmarks focus on capturing the performance of each level of the memory hierarchy, along with thread-level parallelism, instruction-level parallelism and explicit SIMD parallelism, measured in the context of the compilers and run-time environments. We also measure sustained PCIe throughput with four GPU memory managed mechanisms. By combining results from the architecture characterization with the Roofline model based solely on architectural specifications, this work offers insights for performance prediction of current and future architectures and their software systems. To that end, we instrument three applications and plot their resultant performance on the corresponding Roofline model when run on a Blue Gene/Q architecture.

97 MATHEMATICS AND COMPUTING↗

MUST: A Scalable Approach to Runtime Error Detection in MPI Programs

The Message-Passing Interface (MPI) is large and complex. Therefore, programming MPI is error prone. Several MPI runtime correctness tools address classes of usage errors, such as deadlocks or nonportable constructs. To our knowledge none of these tools scales to more than about 100 processes. However, some of the current HPC systems use more than 100,000 cores and future systems are expected to use far more. Since errors often depend on the task count used, we need correctness tools that scale to the full system size. We present a novel framework for scalable MPI correctness tools to address this need. Our fine-grained, module-based approach supports rapid prototyping and allows correctness tools built upon it to adapt to different architectures and use cases. The design uses PnMPI to instantiate a tool from a set of individual modules. We present an overview of our design, along with first performance results for a proof of concept implementation.

99 GENERAL AND MISCELLANEOUS↗

ViennaX: a parallel plugin execution framework for scientific computing

We present the free open source plugin execution framework ViennaX for modularizing and parallelizing scientific simulations. In general, functionality is abstracted by the notion of a task, which is implemented as a plugin. The plugin system facilitates the utilization of both, already available functionality as well as new implementations. Each task can define arbitrary data dependencies which are used by ViennaX to build a task graph. The framework supports the execution of this dependence graph based on the message passing interface in either a serial or a parallel fashion. The applied modular approach allows for defining highly flexible simulations, as plugins can be easily exchanged. The framework's general design as well as implementation details are discussed. Applications based on the Mandelbrot set and the solution of a partial differential equation are investigated, and performance results are shown.

Weinbub, Josef↗

DyG-DPCD: A Distributed Parallel Community Detection Algorithm for Large-Scale Dynamic Graphs

Dynamic (Temporal) graphs capture the valuable evolution of real-world systems, from the continuously evolving patterns of social interactions and genetic pathways to the dynamic fluctuations of economic forces. Detecting communities for such evolving networks poses unique challenges. Detecting and analyzing the evolution of communities within dynamic graphs unlocks valuable insights into the underlying structural and temporal patterns of real-world systems. However, the sheer volume of modern graph data and the inherent complexity of the temporal dimension pose significant challenges to scalable community detection algorithms. Addressing this gap, our work explores the limited landscape of scalable distributed-memory parallel methods specifically designed for dynamic network community detection. We propose a novel parallel algorithm, DyG-DPCD (Dynamic Graph Distributed Parallel Community Detection), to detect communities in dynamic networks using the Message Passing Interface (MPI) framework. We present a vertex-centric approach, allowing us to detect communities through local optimization. Furthermore, we enhance our baseline algorithm by incorporating three heuristics, which improve the algorithm’s performance significantly while maintaining the quality of the solutions. We demonstrate the efficiency of our algorithm by experimenting on several real-world large-scale networks with hundreds of millions of edges spanning diverse domains. Notably, DyG-DPCD achieves speedups between 25× and 30× for large networks that we experimented on using NERSC compute nodes. In conclusion, our algorithm outperforms the STINGER parallel re-agglomeration algorithm by 30×.

97 MATHEMATICS AND COMPUTING↗

Designing Energy Efficient Communication Runtime Systems: A View from PGAS Models

As the march to the Exascale computing gains momentum, energy consumption of supercomputers has emerged to be the critical roadblock. While architectural innovations are imperative in achieving computing of this scale, it is largely dependent on the systems software to leverage the architectural innovations. Parallel applications in many computationally intensive domains have been designed to leverage these supercomputers, with legacy two-sided communication semantics using Message Passing Interface. At the same time, Partitioned Global Address Space Models are being designed which provide global address space abstractions and one-sided communication for exploiting data locality and communication optimizations. PGAS models rely on one-sided communication runtime systems for leveraging high-speed networks to achieve best possible performance. In this paper, we present a design for Power Aware One-Sided Communication Library - PASCoL. The proposed design detects communication slack, leverages Dynamic Voltage and Frequency Scaling (DVFS) and Interrupt driven execution to exploit the detected slack for energy efficiency. We implement our design and evaluate it using synthetic benchmarks for one-sided communication primitives, Put, Get and Accumulate and uniformly non-contiguous data transfers. Our performance evaluation indicates that we can achieve significant reduction in energy consumption without performance loss on multiple one-sided communication primitives. The achieved results are close to the theoretical peak available with the experimental test bed.

Vishnu, Abhinav↗

OpenMP and MPI implementations of an elasto-viscoplastic fast Fourier transform-based micromechanical solver for fast crystal plasticity modeling

Here, we explore several parallel implementations of an elasto-viscoplastic fast Fourier transform (EVPFFT) model using Message Passing Interface (MPI), OpenMP, and a hybrid of MPI and OpenMP to efficiently predict micromechanical response of polycrystals. Performance studies using EVPFFT are performed based on domain decomposition over voxels of a periodic cell, which is a representative volume element (RVE) of polycrystalline copper. We begin by parallelizing the computationally intensive Newton–Raphson (NR) single crystal solver within EVPFFT. Next, we compare the performance of the serial and parallel FFTW (Fastest Fourier Transform in the West) using OpenMP (OpenMP-FFTW) and MPI (MPI-FFTW) with the original Numerical Recipes-based FOURN routine within EVPFFT. In the parallel environment, we find that the FFT calculations are best performed using the MPI version of FFTW. Finally, the remainder of the code, except read/write subroutines, is parallelized. Significant speedups of the original EVPFFT model are achieved using MPI on shared memory multicore workstations. Furthermore, results achieved on a distributed memory Cray supercomputer show promising strong and weak scalability and in some cases even super scalability for the single crystal NR solver in EVPFFT. MPI-FFTW also scales perfectly for microstructure RVEs larger than 64 3 FFT voxels. For example, the MPI-EVPFFT parallel version of the code accelerates the simulations for approximately two orders of magnitude using 64 cores over the old serial code for an RVE size of 128 3 . Finally, the parallel EVPFFT code developed in this work can run massive voxel-based microstructural RVEs taking the advantages of thousands of logical cores provided by more advanced clusters.

42 ENGINEERING↗