Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed parallel programming”

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
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Axom

Axom is an open-source library of "building block" software components that provide core infrastructure capabilities for HPC applications. Such capabilities include: input file parsing and verification for various file formats, format specification for material shape input to multi-material simulations and tools for placing those shapes on meshes, parallel distributed coordination of diagnostic messages, geometric primitives, spatial queries and spatial search acceleration data structures, in memory key-value data store for managing simulation data and parallel file I/O. Axom components are used widely across LLNL Advanced Simulation and Computing program applications. It is also part of the LLNL institutionally-supported RADIUSS project, which promotes and funds its adoption by projects across LLNL.

Zagaris, George↗
DOE OSTI2015

Costa - Introduction to 2015 Annual Report

In parallel with Sandia National Laboratories having two major locations (NM and CA), along with a number of smaller facilities across the nation, so too is the distribution of scientific, engineering and computing resources. As a part of Sandia’s Institutional Computing Program, CA site-based Sandia computer scientists and engineers have been providing mission and research staff with local CA resident expertise on computing options while also focusing on two growing high performance computing research problems. The first is how to increase system resilience to failure, as machines grow larger, more complex and heterogeneous. The second is how to ensure that computer hardware and configurations are optimized for specialized data analytical mission needs within the overall Sandia computing environment, including the HPC subenvironment. All of these activities support the larger Sandia effort in accelerating development and integration of high performance computing into national security missions. Sandia continues to both promote national R&D objectives, including the recent Presidential Executive Order establishing the National Strategic Computing Initiative and work to ensure that the full range of computing services and capabilities are available for all mission responsibilities, from national security to energy to homeland defense.

97 MATHEMATICS AND COMPUTING↗
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 OSTI2023

Reinforcement Learning for Load-balanced Parallel Particle Tracing

We explore an online reinforcement learning (RL) paradigm to dynamically optimize parallel particle tracing performance in distributed-memory systems. Our method combines three novel components: (1) a work donation algorithm, (2) a high-order workload estimation model, and (3) a communication cost model. First, we design an RL-based work donation algorithm. Our algorithm monitors workloads of processes and creates RL agents to donate data blocks and particles from high-workload processes to low-workload processes to minimize program execution time. The agents learn the donation strategy on the fly based on reward and cost functions designed to consider processes' workload changes and data transfer costs of donation actions. Second, we propose a workload estimation model, helping RL agents estimate the workload distribution of processes in future computations. Third, we design a communication cost model that considers both block and particle data exchange costs, helping RL agents make effective decisions with minimized communication costs. We demonstrate that our algorithm adapts to different flow behaviors in large-scale fluid dynamics, ocean, and weather simulation data. Our algorithm improves parallel particle tracing performance in terms of parallel efficiency, load balance, and costs of I/O and communication for evaluations with up to 16,384 processors.

Distributed and parallel particle tracing↗
DOE OSTI2022

Modular Hybrid Solid State Transformer (H-SST) for Next Generation Flexible and Adaptable Large Power Transformer (LPT) (Final Technical Report)

A 500 kVA HSST, the world first of such system, has been successfully developed. The HSST has the capability to regulate the power flow of the transmission or distribution power lines, or provide voltage regulation. The HSST design is modular therefore it has the ability to be connected in series of parallel to form a large power transformer with smart controllability. The project has enabled a high TRL level prototype be demonstrated at University of Texas at Austin which can lead to additional commercialization and demonstration opportunities. A large number of students were trained under this program that will benefit the power grid industry. A number of publications are published at well known international conferences and journals, further accomplish the mission of knowledge development and dissemination.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2014

Final Report for Project DE-FC02-06ER25755 [Pmodels2]

In this report, we describe the research accomplished by the OSU team under the Pmodels2 project. The team has worked on various angles: designing high performance MPI implementations on modern networking technologies (Mellanox InfiniBand (including the new ConnectX2 architecture and Quad Data Rate), QLogic InfiniPath, the emerging 10GigE/iWARP and RDMA over Converged Enhanced Ethernet (RoCE) and Obsidian IB-WAN), studying MPI scalability issues for multi-thousand node clusters using XRC transport, scalable job start-up, dynamic process management support, efficient one-sided communication, protocol offloading and designing scalable collective communication libraries for emerging multi-core architectures. New designs conforming to the Argonne’s Nemesis interface have also been carried out. All of these above solutions have been integrated into the open-source MVAPICH/MVAPICH2 software. This software is currently being used by more than 2,100 organizations worldwide (in 71 countries). As of January ’14, more than 200,000 downloads have taken place from the OSU Web site. In addition, many InfiniBand vendors, server vendors, system integrators and Linux distributors have been incorporating MVAPICH/MVAPICH2 into their software stacks and distributing it. Several InfiniBand systems using MVAPICH/MVAPICH2 have obtained positions in the TOP500 ranking of supercomputers in the world. The latest November ’13 ranking include the following systems: 7th ranked Stampede system at TACC with 462,462 cores; 11th ranked Tsubame 2.5 system at Tokyo Institute of Technology with 74,358 cores; 16th ranked Pleiades system at NASA with 81,920 cores; Work on PGAS models has proceeded on multiple directions. The Scioto framework, which supports task-parallelism in one-sided and global-view parallel programming, has been extended to allow multi-processor tasks that are executed by processor groups. A quantum Monte Carlo application is being ported onto the extended Scioto framework. A public release of Global Trees (GT) has been made, along with the Global Chunks (GC) framework on which GT is built. The Global Chunks (GC) layer is also being used as the basis for the development of a higher level Global Graphs (GG) layer. The Global Graphs (GG) system will provide a global address space view of distributed graph data structures on distributed memory systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

On Eliminating Synchronous Communication in Molecular Simulations to Improve Scalability

Molecular dynamics simulation, as a complementary tool to experimentation, has become an important methodology for the understanding and design of molecular systems as it provides access to properties that are difficult, impossible or prohibitively expensive to obtain experimentally. Many of the available software packages have been parallelized to take advantage of modern massively concurrent processing resources. The challenge in achieving parallel efficiency is commonly attributed to the fact that molecular dynamics algorithms are communication intensive. This paper illustrates how an appropriately chosen data distribution and asynchronous one-sided communication approach can be used to effectively deal with the data movement within the Global Arrays/ARMCI programming model framework. A new put_notify capability is presented here, allowing the implementation of the molecular dynamics algorithm without any explicit global or local synchronization or global data reduction operations. In addition, this push-data model is shown to very effectively allow hiding data communication behind computation. Rather than data movement or explicit global reductions, the implicit synchronization of the algorithm becomes the primary challenge for scalability. Without any explicit synchronous operations, the scalability of molecular simulations is shown to depend only on the ability to evenly balance computational load.

Straatsma, TP↗
DOE OSTI2017

ATDM LANL FleCSI: Topology and Execution Framework

FleCSI is a compile-time configurable C++ framework designed to support multi-physics application development. As such, FleCSI attempts to provide a very general set of infrastructure design patterns that can be specialized and extended to suit the needs of a broad variety of solver and data requirements. This means that FleCSI is potentially useful to many different ECP projects. Current support includes multidimensional mesh topology, mesh geometry, and mesh adjacency information, n-dimensional hashed-tree data structures, graph partitioning interfaces, and dependency closures (to identify data dependencies between distributed-memory address spaces). FleCSI introduces a functional programming model with control, execution, and data abstractions that are consistent with state-of-the-art task-based runtimes such as Legion and Charm++. The model also provides support for fine-grained, data-parallel execution with backend support for runtimes such as OpenMP and C++17. The FleCSI abstraction layer provides the developer with insulation from the underlying runtimes, while allowing support for multiple runtime systems, including conventional models like asynchronous MPI. The intent is to give developers a concrete set of user-friendly programming tools that can be used now, while allowing flexibility in choosing runtime implementations and optimizations that can be applied to architectures and runtimes that arise in the future. This project is essential to the ECP Ristra Next-Generation Code project, part of ASC ATDM, because it provides a hierarchically parallel programming model that is consistent with the design of modern system architectures, but which allows for the straightforward expression of algorithmic parallelism in a portably performant manner.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

New Results on Communication- and Memory-Aware Load Balancing Model and Algorithms

While load balancing in distributed-memory computing has been well-studied, we present an innovative approach to this problem: a unified, reduced-order model that combines three key components to describe “work” in a distributed system: computation, communication, and memory. Our model enables an optimizer to explore complex tradeoffs in task placement, such as augmented parallelism, at the expense of data replication increasing memory usage. We propose a fully distributed, heuristic-based load balancing optimization algorithm, and demonstrate that it quickly finds close-to-optimal solutions. We formalize the complex optimization problem as a mixed-integer linear program, and compare it to our strategy. Finally, we show that when applied to an electromagnetics code, our approach obtains up to 2.3x speedups for the imbalanced execution.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Trigger-based Incremental Data Processing with Unified Sync and Async Model

In recent years, more and more applications in the cloud have needs to process large-scale on-line datasets, which evolve over time as new entries are added and existing entries are modified. Several programming frameworks, such as Percolator and Oolong, are proposed for such incremental data processing and can achieve efficient processing with an event-driven abstraction. However, these frameworks are inherently asynchronous, leaving the heavy burden of managing synchronization to applications' developers, which further significantly restricts their usabilities. In this study, we propose a trigger-based incremental computing framework in the cloud, called Domino, with both synchronous and asynchronous mechanisms to coordinate parallel triggers. With this new framework, both synchronous and asynchronous applications can be seamlessly developed. Use cases and extensive evaluation results confirm that it can deliver sufficient performance, and also is easy to use for incremental applications in large-scale distributed computing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Synchronization-Avoiding Graph Algorithms

Because they were developed for optimal sequential complexity, classical graph algorithms as found in textbooks have strictly-defined orders of operations. Enforcing a prescribed order of operations, or even an approximate order, in a distributed memory setting requires significant amounts of synchronization, which in turn can severely limit scalability. As a result, new algorithms are typically required to achieve scalable performance, even for solving well-known graph problems. Yet, even in these cases, parallel graph algorithms are written according to parallel programming models that evolved for, e.g., scientific computing, and that still have inherent, and scalability-limiting, amounts of synchronization. In this paper we present a new approach to parallel graph algorithms: synchronization-avoiding algorithms. To eliminate synchronization and its associated overhead, synchronization-avoiding algorithms perform work in an unordered and fully asynchronous fashion in such a way that the result is constantly refined toward its final state. “Wasted” work is minimized by locally prioritizing tasks using problem-dependent task utility metrics. We classify algorithms for graph applications into two broad categories: algorithms with monotonic updates (which evince global synchronization) and algorithms with nonmonotonic updates (which evince vertex-centric synchronization). We apply our approach to both classes and develop novel, synchronization-avoiding algorithms for solving exemplar problems: SSSP and connected components for the former, graph coloring for the latter. We demonstrate that eliminating synchronization in conjunction with effective scheduling policies and optimizations in the runtime results in improved scalability for both classes of algorithms.

Firoz, Jesun S.↗
DOE OSTI2019

Use of Neutron Flux Calculated by Shift in a Grizzly Reactor Pressure Vessel Fracture Simulation

The reactor pressure vessel (RPV) plays the critical role of containing the reactor and coolant in a light water reactor nuclear power plant, and must be able to safely perform this function under a variety of normal and off-normal transient loading conditions. The RPV is subjected to harsh environmental conditions (elevated temperature and high neutron flux) whenever the plant is operating, which leads to embrittlement of the steel over time. RPVs contain populations of flaws introduced during the manufacturing process, and the primary safety concern is that during a transient event, a fracture could initiate at the site of one of these flaws and lead to rupture of the RPV. RPVs are very rugged structures, and the probability of this occurring is very low, but as the RPV's steel becomes embrittled over time, its resistance to fracture decreases and this becomes more likely. Assessing the likelihood of RPV fracture during a transient event involves performing a probabilistic fracture mechanics analysis, in which random sampling procedures are used to assess the probability of fracture of a population of flaws. Grizzly offers unique capabilities for modeling the RPV using 1D, 2D, or 3D representations, taking advantage of parallel computing for this type of analysis. One of the important inputs to Grizzly is the spatial distribution of the neutron fluence, which has a large impact on the material embrittlement, and should thus be characterized as accurately as possible. The VERA core simulator developed by the CASL program takes advantage of high performance computing resources to perform high-fidelity multiphysics simulations of reactor physics, including the effects of neutron transport, fuel performance, and thermal hydraulics. Recently a capability was developed to also evaluate neutron fluxes away from the reactor core through the Shift code. This allows for a detailed 3D map of the spatial variation of the fluence in components such as the RPV to be computed, which is ideal for use with Grizzly. This report documents a procedure developed to use the fluences computed in Shift in a Grizzly simulation, which allows for improved realisim in the prediction of the effects of environmental exposure on propensity for reactor pressure vessel fracture. This capability demonstrated on RPVs with multiple types of configurations.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2023

A Sparse Distributed Gigascale Resolution Material Point Method

In this paper, we present a four-layer distributed simulation system and its adaptation to the Material Point Method (MPM). The system is built upon a performance portable C++ programming model targeting major High-Performance-Computing (HPC) platforms. A key ingredient of our system is a hierarchical block-tile-cell sparse grid data structure that is distributable to an arbitrary number of Message Passing Interface (MPI) ranks. We additionally propose strategies for efficient dynamic load balance optimization to maximize the efficiency of MPI tasks. Our simulation pipeline can easily switch among backend programming models, including OpenMP and CUDA, and can be effortlessly dispatched onto supercomputers and the cloud. Finally, we construct benchmark experiments and ablation studies on supercomputers and consumer workstations in a local network to evaluate the scalability and load balancing criteria. We demonstrate massively parallel, highly scalable, and gigascale resolution MPM simulations of up to 1.01 billion particles for less than 323.25 seconds per frame with 8 OpenSSH-connected workstations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

A Fine-grained Asynchronous Bulk Synchronous parallelism model for PGAS applications

The Partitioned Global Address Space (PGAS) model is well suited for executing irregular applications on cluster-based systems, due to its efficient support for short, one-sided messages. Separately, the actor model has been gaining popularity as a productive asynchronous message-passing approach for distributed objects in enterprise and cloud computing platforms, typically implemented in languages such as Erlang, Scala or Rust. To the best of our knowledge, there has been no past work on using the actor model to deliver both productivity and scalability to irregular PGAS applications with large number of small messages. In this paper, we introduce a new programming system for PGAS applications, in which point-to-point remote operations can be expressed as fine-grained asynchronous actor messages. In our approach, the programmer does not need to worry about programming complexities related to message aggregation and termination detection. Our approach can be viewed as extending the classical Bulk Synchronous Parallelism model with fine-grained asynchronous communications within a phase or superstep. Here, we believe that our approach offers a desirable point in the productivity-performance space for PGAS applications, with more scalable performance and higher productivity relative to past approaches. Specifically, for seven irregular mini-applications from the Bale Kernels and three graph kernels executed using 2048 cores in the NERSC Cori system, our approach shows geometric mean performance improvements of ≥ 20X relative to standard PGAS versions (UPC and OpenSHMEM) while maintaining comparable productivity to those versions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2013

THE SPECTRAL ENERGY DISTRIBUTIONS OF z ∼ 8 GALAXIES FROM THE IRAC ULTRA DEEP FIELDS: EMISSION LINES, STELLAR MASSES, AND SPECIFIC STAR FORMATION RATES AT 650 MYR

Using new ultradeep Spitzer/InfraRed Array Camera (IRAC) photometry from the IRAC Ultra Deep Field program, we investigate the stellar populations of a sample of 63 Y-dropout galaxy candidates at z ∼ 8, only 650 Myr after the big bang. The sources are selected from HST/ACS+WFC3/IR data over the Hubble Ultra Deep Field (HUDF), two HUDF parallel fields, and wide area data over the CANDELS/GOODS-South. The new Spitzer/IRAC data increase the coverage in [3.6] and [4.5] to ∼120h over the HUDF reaching depths of ∼28 (AB,1σ). The improved depth and inclusion of brighter candidates result in direct ≥3σ InfraRed Array Camera (IRAC) detections of 20/63 sources, of which 11/63 are detected at ≥5σ. The average [3.6]-[4.5] colors of IRAC detected galaxies at z ∼ 8 are markedly redder than those at z ∼ 7, observed only 130 Myr later. The simplest explanation is that we witness strong rest-frame optical emission lines (in particular [O III] λλ4959, 5007 + Hβ) moving through the IRAC bandpasses with redshift. Assuming that the average rest-frame spectrum is the same at both z ∼ 7 and z ∼ 8 we estimate a rest-frame equivalent width of contributing 0.56{sup +0.16}{sub -0.11} mag to the [4.5] filter at z ∼ 8. The corresponding W{sub Hα}=430{sup +160}{sub -110} Å implies an average specific star formation rate of sSFR=11{sub -5}{sup +11} Gyr{sup –1} and a stellar population age of 100{sub -50}{sup +100} Myr. Correcting the spectral energy distribution for the contribution of emission lines lowers the average best-fit stellar masses and mass-to-light ratios by ∼3 ×, decreasing the integrated stellar mass density to ρ{sup *}(z=8,M{sub UV}<-18)=0.6{sup +0.4}{sub -0.3}×10{sup 6} M{sub sun} Mpc{sup –3}.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

Application Characterization at Scale: Lessons learned from developing a distributed Open Community Runtime system for High Performance Computing

Since 2012, the U.S. Department of Energy’s X-Stack program has been developing solutions including runtime systems, programming models, languages, compilers, and tools for the Exascale system software to address crucial performance and power requirements. Fine grain programming models and runtime systems show a great potential to efficiently utilize the underlying hardware. Thus, they are essential to many X-Stack efforts. An abundant amount of small tasks can better utilize the vast parallelism available on current and future machines. Moreover, finer tasks can recover faster and adapt better, due to a decrease in state and control. Nevertheless, current applications have been written to exploit old paradigms (such as Communicating Sequential Processor and Bulk Synchronous Parallel processing). To fully utilize the advantages of these new systems, applications need to be adapted to these new paradigms. As part of the applications’ porting process, in-depth characterization studies, focused on both application characteristics and runtime features, need to take place to fully understand the application performance bottlenecks and how to resolve them. This paper presents a characterization study for a novel high performance runtime system, called the Open Community Runtime, using key HPC kernels as its vehicle. This study has the following contributions: one of the first high performance, fine grain, distributed memory runtime system implementing the OCR standard (version 0.99a); and a characterization study of key HPC kernels in terms of runtime primitives running on both intra and inter node environments. Running on a general purpose cluster, we have found up to 1635x relative speed-up for a parallel tiled Cholesky Kernels on 128 nodes with 16 cores each and a 1864x relative speed-up for a parallel tiled Smith-Waterman kernel on 128 nodes with 30 cores.

Application characterization↗
DOE OSTI2024

Influence of Antarctic and Greenland Continental Shelf Circulation on High‐Latitude Oceans in E3SM

The science objectives of this project are to simulate and understand the impacts of both deep-basin warm-water intrusions and land-ice melt on the continental shelf circulations and sea-ice distributions around the margins of Greenland and Antarctica. As well, the role of subsurface ocean heat from the Atlantic on declining sea-ice cover in the Arctic is explored. Mesoscale processes and fine bathymetry are implicated in cross-shelf property transports around both Greenland and Antarctica. Therefore, we configured and ran an atmospheric reanalysis-forced global ocean/sea-ice simulation on a grid that reduces from 8 km at the Equator to 2 km at the poles (UH8to2) with 60 vertical levels. It was produced using the Energy Exascale Earth System Model ‘‘HiLAT’’ code (E3SMv0-HiLAT) that uses the Parallel Ocean Program (POP) and CICE5 as its ocean and sea-ice components, respectively. Two main UH8to2 simulations were carried out: one for 1975-2009 and the other for July 2016-2020 after it was initialized from a 1/25° data-assimilative ocean/sea-ice prediction system ocean/sea-ice state. The UH8to2 is not coupled to an active land-ice model. Rather, land-ice melt is represented by observationally informed freshwater fluxes (FWFs). Short (multi-year) UH8to2 simulations were conducted to understand sensitivities when Greenland ice sheet (GrIS) melt is released only at the ocean surface or when it is distributed over the upper water column in accordance with fjord melt plume behavior; these cases were compared with a no GrIS melt case. West Greenland continental shelf currents were fastest in the vertical distribution case and an increase in baroclinic conversion at the shelf break associated with increased eddy kinetic energy was found relative to the surface release case. Further, salinity is lower and meltwater volume greater in the eastern Labrador Sea in the vertical distribution case. For the Arctic, the veracity of the UH8to2 was evaluated for 2017-2020 using available observations. Simulated seasonal sea-ice thickness and concentration are realistic, but the ice is unrealistically thin in the central and eastern Arctic in the fall. Comparisons of vertical sections of ocean temperature, salinity, and buoyancy collected from Ice-Tethered Profilers (ITPs) in the eastern Arctic in the fall and winter of 2019/2020 and co-located/concurrent UH8to2 fields show the stratification over the top 100 m of the water column is too low in the model, the simulated mixed layer too deep, and the simulated subsurface Atlantic Water (AW) too warm; these biases may contribute to the sea-ice biases. A model intercomparison study using the UH8to2 and a forced 1/25° regional Arctic ocean/sea-ice (uses the HYbrid Coordinate Ocean Model and CICE5) simulation further investigates the relationship between AW and sea-ice in the eastern Arctic. The models show a mesoscale-rich pulse of Atlantic Water extending into the eastern basin that reaches maximum intensity in late winter of 2018, after which it decreases in strength. Concurrent and co-located sea-ice melt or the inhibition of sea-ice growth is seen and is attributed to halocline mesoscale eddies doming into the mixed layer with convection bringing this heat into the vicinity of the sea-ice.

58 GEOSCIENCES↗