Search NASA⌕ Search

SEARCH · Search NASA

Results for “Global Arrays”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Global Arrays

Global Arrays (GA) is a distributed-memory programming model that allows for shared-memory-style programming combined with one-sided communication, to create a set of tools that combine high performance with ease-of-use. GA exposes a relatively straightforward programming abstraction, while supporting fully-distributed data structures, locality of reference, and high-performance communication. GA was originally formulated in the early 1990’s to provide a communication layer for the Northwest Chemistry (NWChem) suite of chemistry modeling codes that was being developed concurrently.

global arrays↗
DOE OSTI2011

Automatic Parallelization of Numerical Python Applications using the Global Arrays Toolkit

Global Arrays is a software system from Pacific Northwest National Laboratory that enables an efficient, portable, and parallel shared-memory programming interface to manipulate distributed dense arrays. The NumPy module is the de facto standard for numerical calculation in the Python programming language, a language whose use is growing rapidly in the scientific and engineering communities. NumPy provides a powerful N-dimensional array class as well as other scientific computing capabilities. However, like the majority of the core Python modules, NumPy is inherently serial. Using a combination of Global Arrays and NumPy, we have reimplemented NumPy as a distributed drop-in replacement called Global Arrays in NumPy (GAiN). Serial NumPy applications can become parallel, scalable GAiN applications with only minor source code changes. Scalability studies of several different GAiN applications will be presented showing the utility of developing serial NumPy codes which can later run on more capable clusters or supercomputers.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2024

Distributed-Memory Sparse Deep Neural Network Inference Using Global Arrays

Partitioned Global Address Space (PGAS) models exhibit tremendous promise in developing efficient and productive distributed-memory parallel applications. They have been used extensively in scientific computations due to conveniently offering a ``shared-memory''-like model and convenient interfaces that separate communication with synchronization. Traditionally, PGAS communication models have been applied to dense/contiguously distributed data, but most modern applications depict varied levels of sparsity. Existing PGAS models require certain adaptations to support distributed sparse computations, since associated computations often require matrix arithmetic, in addition to data movement. The Global Arrays toolkit from Pacific Northwest National Laboratory (PNNL) is one of the earliest PGAS models to combine one-sided data communication and distributed matrix operations and is still used in the popular NWChem quantum chemistry suite. Recently, we have expanded the Global Arrays toolkit to support common sparse operations, like sparse matrix-dense matrix multiplies (SpMM), sparse matrix-sparse matrix multiplication (SpGEMM) and Sampled Dense-Dense Matrix Multiplication (SDDMM). As it turns out, these operations are the bedrock of sparse Deep Learning (DL); sparse deep neural networks and Graph Neural Networks (GNNs) have gained increasing attention recently in achieving speedups on training and inference with reduced memory footprints. Unlike scientific applications in High Performance Computing (HPC), modern (distributed-memory capable) DL toolkits often rely on non-standardized and closed-source vendor software optimizations, creating challenges in software-hardware co-design at scale. Our goal is to support a variety of distributed-memory sparse matrix operations and helper functions in the newly created Sparse Global Arrays (SGA), such that it is possible to build portable and productive Machine Learning scenarios for algorithm/software and hardware codesign purposes. Contemporary data-parallel schemes for training/inference are undergoing a major overhaul since model replication limits scalability and causes resource inefficiencies. As such, we have adopted tensor parallelism in decomposing the model and inputs, to mitigate memory issues. Current implementation is built on top of MPI and uses CPUs to maximize the portability across the platforms.

Distributed computing, machine learning↗
DOE OSTI2013

Efficient Memory Access with NumPy Global Arrays using Local Memory Access

This paper discusses the work completed working with Global Arrays of data on distributed multi-computer systems and improving their performance. The tasks completed were done at Pacific Northwest National Laboratory in the Science Undergrad Laboratory Internship program in the summer of 2013 for the Data Intensive Computing Group in the Fundamental and Computational Sciences DIrectorate. This work was done on the Global Arrays Toolkit developed by this group. This toolkit is an interface for programmers to more easily create arrays of data on networks of computers. This is useful because scientific computation is often done on large amounts of data sometimes so large that individual computers cannot hold all of it. This data is held in array form and can best be processed on supercomputers which often consist of a network of individual computers doing their computation in parallel. One major challenge for this sort of programming is that operations on arrays on multiple computers is very complex and an interface is needed so that these arrays seem like they are on a single computer. This is what global arrays does. The work done here is to use more efficient operations on that data that requires less copying of data to be completed. This saves a lot of time because copying data on many different computers is time intensive. The way this challenge was solved is when data to be operated on with binary operations are on the same computer, they are not copied when they are accessed. When they are on separate computers, only one set is copied when accessed. This saves time because of less copying done although more data access operations were done.

Daily, Jeffrey A.↗
DOE OSTI2019

Parallel Distributed Hydrology Soil Vegetation Model (DHSVM) using global arrays

The Distributed Hydrology Soil Vegetation Model (DHSVM) code was parallelized for distributed memory computers using Global Arrays (GA). To analyze parallel performance, DHSVM was used to simulate two significantly sized river basins located in northwest continental United States and southwest Canada at 90~m resolution: (1) Clearwater (25,000~km2) and (2) Columbia (668,000~km2). Meteorological forcing applied to both basins was dynamically down-scaled from a global circulation model using the Weather Research and Forecasting (WRF) model and read into DHSVM as 2D maps each time step. We report parallel code speed up was significant. Run times for 1-year simulations were reduced by an order of magnitude for both test basins. A maximum speed up of 105 was attained with 480 processors while simulating the Columbia basin. Speed up was limited by input dominated tasks, particularly the input of meteorological forcing data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2012

Supporting the Global Arrays PGAS Model Using MPI One-Sided Communication

The industry-standard Message Passing Interface (MPI) provides one-sided communication functionality and is available on virtually every parallel computing system. However, it is believed that MPI’s one-sided model is not rich enough to support higher-level global address space parallel programming models. We present the first successful application of MPI one-sided communication as a runtime system for a PGAS model, Global Arrays (GA). This work has an immediate impact on users of GA applications, such as NWChem, who often must wait several months to a year or more before GA becomes available on a new architecture. We explore challenges present in the application of MPI-2 to PGAS models and motivate new features in the upcoming MPI-3 standard. The performance of our system is evaluated on several popular high-performance computing architectures through communication benchmarking and application benchmarking using the NWChem computational chemistry suite.

ARMCI↗
DOE OSTI2019

Accelerating the Global Arrays ComEx Runtime using Multiple Progress Ranks

Partitioned Global Address Space (PGAS) models are part of system software that are being designed to support communication runtime for exascale applications. Suitability of using MPI as a scalable PGAS communication subsystem has been shown as a viable option due to its standardization and higher performance. We used MPI two-sided semantics with a combination of automatic and user defined splitting of MPI communicators to achieve asynchronous progress. Our implementation can make use of multiple asynchronous progress ranks (PR) per node that can be mapped to the computing architecture of a node in a distributed cluster. We are able to show significant speed up of over 2.0X and scaling of communication bound computational chemistry application over 1024 nodes of state- of-the-art HPC clusters. Our results show that while running a communication bound application workload on a certain number of cluster nodes, a use of optimum number of ranks dedicated for communication may be used to achieve asynchronous communication progress and obtain highest performance.

PGAS, global-arrays, NWCHEM, NWChemEx, HPC, GPGPU↗
DOE OSTI2012

Global Futures: a multithreaded execution model for Global Arrays-based applications

We present Global Futures (GF), an execution model extension to Global Arrays, which is based on a PGAS-compatible Active Message-based paradigm. We describe the design and implementation of Global Futures and illustrate its use in a computational chemistry application benchmark (Hartree-Fock matrix construction using the Self-Consistent Field method). Our results show how we used GF to increase the scalability of the Hartree-Fock matrix build to up to 6,144 cores of an Infiniband cluster. We also show how GF's multithreaded execution has comparable performance to the traditional process-based SPMD model.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2013

Theoretical analysis of phase locking in an array of globally coupled lasers

A model of an array of globally coupled fibre lasers, with the same fraction of the total output beam power injected into each laser, is considered. Phase self-locking of the laser array makes it possible to increase the brightness of the total output beam without any devices for controlling the phases of output beams, which significantly complicate the laser system. The spread of the laser optical lengths is several hundreds of wavelengths (or even more); within the theory of hollow cavities, this spread should lead to a fast decrease in the total power with an increase in the number of lasers. The presence of the active medium may reduce this drop to a great extent due to the self-tuning of the laser array radiation wavelength to a value providing a maximum gain for the array lasing mode. The optical length of each element is assumed to be random. The increase in the phase-locking efficiency due to the gain saturation is explained based on the probabilistic approach. An iterative procedure is developed to find the array output power in the presence of steady-state phase locking. Calculations for different values of small-signal gain and the output-power fraction spent on global coupling are performed. It is shown that, when this fraction amounts to ∼20 % – 30 %, phase locking of up to 20 fibre lasers can be implemented with an efficiency as high as 70 %. (control of laser radiation parameters)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2014

Universal Common Communication Substrate (UCCS) Specification; Universal Common Communication Substrate (UCCS) Implementation

Universal Common Communication Substrate (UCCS) is a low-level communication substrate that exposes high-performance communication primitives, while providing network interoperability. It is intended to support multiple upper layer protocol (ULPs) or programming models including SHMEM,UPC,Titanium,Co-Array Fortran,Global Arrays,MPI,GASNet, and File I/O. it provides various communication operations including one-sided and two-sided point-to-point, collectives, and remote atomic operations. In addition to operations for ULPs, it provides an out-of-band communication channel required typically required to wire-up communication libraries.

Source record↗
DOE OSTI2016

Phase-locked laser array through global antenna mutual coupling

Here, phase locking of an array of lasers is a highly effective way in beam shaping, to increase the output power, and to reduce lasing threshold. In this work, we present a novel phase-locking mechanism based on "antenna mutual coupling" wherein laser elements interact through far-field radiations with definite phase relations. This allows long-range global coupling among array elements to achieve robust 2-dimensional phase-locked laser array. The new scheme is ideal for lasers with deep sub-wavelength confined cavity such as nanolasers, where the divergent beam pattern could be used to form strong coupling among elements in the array. We experimentally demonstrated such a scheme using sub-wavelength short-cavity surface-emitting lasers at terahertz frequency. More than 37 laser elements are phase-locked to each other, delivering up to 6.5 mW single-mode radiations at ~3 terahertz, with maximum 450-mW/A slope efficiency and near diffraction limit beam divergence.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

Exploring versioned distributed arrays for resilience in scientific applications: Global view resilience

Exascale studies project reliability challenges for future HPC systems. We present the Global View Resilience (GVR) system, a library for portable resilience. GVR begins with a subset of the Global Arrays interface, and adds new capabilities to create versions, name versions, and compute on version data. Applications can focus versioning where and when it is most productive, and customize for each application structure independently. This control is portable, and its embedding in application source makes it natural to express and easy to maintain. The ability to name multiple versions and “partially materialize” them efficiently makes ambitious forward-recovery based on “data slices” across versions or data structures both easy to express and efficient. Using several large applications (OpenMC, preconditioned conjugate gradient (PCG) solver, ddcMD, and Chombo), we evaluate the programming effort to add resilience. The required changes are small (< 2% lines of code (LOC)), localized and machine-independent, and perhaps most important, require no software architecture changes. We also measure the overhead of adding GVR versioning and show that overheads < 2% are generally achieved. This overhead suggests that GVR can be implemented in large-scale codes and support portable error recovery with modest investment and runtime impact. Our results are drawn from both IBM BG/Q and Cray XC30 experiments, demonstrating portability. We also present two case studies of flexible error recovery, illustrating how GVR can be used for multi-version rollback recovery, and several different forward-recovery schemes. GVR’s multi-version enables applications to survive latent errors (silent data corruption) with significant detection latency, and forward recovery can make that recovery extremely efficient. Lastly, our results suggest that GVR is scalable, portable, and efficient. GVR interfaces are flexible, supporting a variety of recovery schemes, and altogether GVR embodies a gentle-slope path to tolerate growing error rates in future extreme-scale systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Global‐view coefficients: a data management solution for parallel quantum Monte Carlo applications

Summary Quantum Monte Carlo (QMC) applications perform simulation with respect to an initial state of the quantum mechanical system, which is often captured by using a cubic B‐spline basis. This representation is stored as a read‐only table of coefficients and accesses to the table are generated at random as part of the Monte Carlo simulation. Current QMC applications, such as QWalk and QMCPACK, replicate this table at every process or node, which limits scalability because increasing the number of processors does not enable larger systems to be run. We present a partitioned global address space approach to transparently managing this data using Global Arrays in a manner that allows the memory of multiple nodes to be aggregated. We develop an automated data management system that significantly reduces communication overheads, enabling new capabilities for QMC codes. Experimental results with QWalk and QMCPACK demonstrate the effectiveness of the data management system. Copyright © 2016 John Wiley & Sons, Ltd.

Niu, Qingpeng↗
DOE OSTI2011

NWChem: scalable parallel computational chemistry

NWChem is a general purpose computational chemistry code specifically designed to run on distributed memory parallel computers. The core functionality of the code focuses on molecular dynamics, Hartree-Fock and density functional theory methods for both plane-wave basis sets as well as Gaussian basis sets, tensor contraction engine based coupled cluster capabilities and combined quantum mechanics/molecular mechanics descriptions. It was realized from the beginning that scalable implementations of these methods required a programming paradigm inherently different from what message passing approaches could offer. In response a global address space library, the Global Array Toolkit, was developed. The programming model it offers is based on using predominantly one-sided communication. This model underpins most of the functionality in NWChem and the power of it is exemplified by the fact that the code scales to tens of thousands of processors. In this paper the core capabilities of NWChem are described as well as their implementation to achieve an efficient computational chemistry code with high parallel scalability. NWChem is a modern, open source, computational chemistry code1 specifically designed for large scale parallel applications2. To meet the challenges of developing efficient, scalable and portable programs of this nature a particular code design was adopted. This code design involved two main features. First of all, the code is build up in a modular fashion so that a large variety of functionality can be integrated easily. Secondly, to facilitate writing complex parallel algorithms the Global Array toolkit was developed. This toolkit allows one to write parallel applications in a shared memory like approach, but offers additional mechanisms to exploit data locality to lower communication overheads. This framework has proven to be very successful in computational chemistry but is applicable to any engineering domain. Within the context created by the features above NWChem has grown into a general purpose computational chemistry code that supports a wide variety of energy expressions and capabilities to calculate properties based there upon. The main energy expressions are classical mechanics force fields, Hartree-Fock and DFT both for finite systems and condensed phase systems, coupled cluster, as well as QM/MM. For most energy expressions single point calculations, geometry optimizations, excited states, and other properties are available. Below we briefly discuss each of the main energy expressions and the critical points involved in scalable implementations thereof.

99 GENERAL AND MISCELLANEOUS↗
DOE OSTI2016

The young and bright type ia supernova ASASSN-14lP: discovery, early-time observations, first-light time, distance to NGC 4666, and progenitor constraints

On 2014 December 9.61, the All-sky Automated Survey for SuperNovae (ASAS-SN or “Assassin”) discovered ASASSN-14lp just ∼2 days after first light using a global array of 14 cm diameter telescopes. ASASSN-14lp went on to become a bright supernova (V = 11.94 mag), second only to SN 2014J for the year. We present prediscovery photometry (with a detection less than a day after first light) and ultraviolet through near-infrared photometric and spectroscopic data covering the rise and fall of ASASSN-14lp for more than 100 days. We find that ASASSN-14lp had a broad light curve (Δm{sub 15}(B)=0.80±0.05), a B-band maximum at 2457015.82 ± 0.03, a rise time of 16.94{sub −0.10}{sup +0.11} days, and moderate host-galaxy extinction (E(B−V){sub host}=0.33±0.06). Using ASASSN-14lp, we derive a distance modulus for NGC 4666 of μ=30.8±0.2, corresponding to a distance of 14.7 ± 1.5 Mpc. However, adding ASASSN-14lp to the calibrating sample of Type Ia supernovae still requires an independent distance to the host galaxy. Finally, using our early-time photometric and spectroscopic observations, we rule out red giant secondaries and, assuming a favorable viewing angle and explosion time, any nondegenerate companion larger than 0.34 R{sub ⊙}.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

Versioned distributed arrays for resilience in scientific applications: Global view resilience

Exascale studies project reliability challenges for future high-performance computing (HPC) systems. We propose the Global View Resilience (GVR) system, a library that enables applications to add resilience in a portable, application-controlled fashion using versioned distributed arrays. We describe GVR’s interfaces to distributed arrays, versioning, and cross-layer error recovery. Using several large applications (OpenMC, the preconditioned conjugate gradient solver PCG, ddcMD, and Chombo), we evaluate the programmer effort to add resilience. The required changes are small (<2% LOC), localized, and machine-independent, requiring no software architecture changes. We also measure the overhead of adding GVR versioning and show that generally overheads <2% are achieved. We conclude that GVR’s interfaces and implementation are flexible and portable and create a gentle-slope path to tolerate growing error rates in future systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2012

Runtime Techniques to Enable a Highly-Scalable Global Address Space Model for Petascale Computing

Over the past decade, the trajectory to the petascale has been built on increased complexity and scale of the underlying parallel architectures. Meanwhile, software developers have struggled to provide tools that maintain the productivity of computational science teams using these new systems. In this regard, Global Address Space (GAS) programming models provide a straightforward and easy to use addressing model, which can lead to improved productivity. However, the scalability of GAS depends directly on the design and implementation of the runtime system on the target petascale distributed-memory architecture. In this paper, we describe the design, implementation, and optimization of the Aggregate Remote Memory Copy Interface (ARMCI) runtime library on the Cray XT5 2.3 PetaFLOPs computer at Oak Ridge National Laboratory. We optimized our implementation with the flow intimation technique that we have introduced in this paper. Our optimized ARMCI implementation improves scalability of both the Global Arrays programming model and a real-world chemistry application—NWChem—from small jobs up through 180,000 cores.

Computer Science↗