Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Parallel computers.”

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 1,063 records · Page 59

Fast non-overlapping Schwarz domain decomposition methods for solving the neutron diffusion equation

Studying numerically the steady state of a nuclear core reactor is expensive, in terms of memory storage and computational time. In order to address both requirements, one can use a domain decomposition method, implemented on a parallel computer. We present here such a method for the mixed neutron diffusion equations, discretized with Raviart–Thomas–Nédélec finite elements. This method is based on the Schwarz iterative algorithm with Robin interface conditions to handle communications. We analyse this method from the continuous point of view to the discrete point of view, and we give some numerical results in a realistic highly heterogeneous 3D configuration. Computations are carried out with the MINOS solver of the APOLLO3® neutronics code.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

PaRSEC: Scalability, flexibility, and hybrid architecture support for task-based applications in ECP

This paper highlights the most significant enhancements made to PaRSEC, a scalable task-based runtime system designed for hybrid machines, during the Exascale Computing Project (ECP). The enhancements focus on expanding the capabilities of PaRSEC to address the evolving landscape of parallel computing. Notable achievements include the integration of support for three major types of accelerators (NVIDIA, AMD, and Intel GPUs), the refinement and increased flexibility of the communication subsystem, and the introduction of new programming interfaces tailored for irregular applications. Additionally, the project resulted in the development of powerful debugging and performance analysis tools aimed at assisting users in understanding and optimizing their applications. We present a comprehensive demonstration of these advancements through a series of benchmarks and applications within ECP and beyond, thereby showcasing the enhanced capabilities of PaRSEC across the diverse architectures within the ECP, providing valuable insights into the runtime system’s adaptability and performance across varied computing environments.

Bouteiller, Aurelien↗

High speed civil transport aerodynamic optimization

This is a report of work in support of the Computational Aerosciences (CAS) element of the Federal HPCC program. Specifically, CFD and aerodynamic optimization are being performed on parallel computers. The long-range goal of this work is to facilitate teraflops-rate multidisciplinary optimization of aerospace vehicles. This year's work is targeted for application to the High Speed Civil Transport (HSCT), one of four CAS grand challenges identified in the HPCC FY 1995 Blue Book. This vehicle is to be a passenger aircraft, with the promise of cutting overseas flight time by more than half. To meet fuel economy, operational costs, environmental impact, noise production, and range requirements, improved design tools are required, and these tools must eventually integrate optimization, external aerodynamics, propulsion, structures, heat transfer, controls, and perhaps other disciplines. The fundamental goal of this project is to contribute to improved design tools for U.S. industry, and thus to the nation's economic competitiveness.

Ryan, James S.↗

XMOS XC-2 Development Board for Mechanical Control and Data Collection

The scanning microwave limb sounder (SMLS) will use technological improvements in low-noise mixers to provide precise data on the Earth s atmospheric composition with high spatial resolution. This project focuses on the design and implementation of a realtime control system needed for airborne engineering tests of the SMLS. The system must coordinate the actuation of optical components using four motors with encoder readback, while collecting synchronized telemetric data from a GPS receiver and 3-axis gyrometric system. A graphical user interface for testing the control system was also designed using Python. Although the system could have been implemented with an FPGA(fieldprogrammable gate array)-based setup, a processor development kit manufactured by XMOS was chosen. The XMOS architecture allows parallel execution of multiple tasks on separate threads, making it ideal for this application. It is easily programmed using XC (a subset of C). The necessary communication interfaces were implemented in software, including Ethernet, with significant cost and time reduction compared to an FPGA-based approach. A simple approach to control the chopper, calibration mirror, and gimbal for the airborne SMLS was needed. The XMOS board allows for multiple threads and real-time data acquisition. The XC-2 development kit is an attractive choice for synchronized, real-time, event-driven applications. The XMOS is based on the transputer microprocessor architecture developed for parallel computing, which is being revamped in this new platform. The XMOS device has multiple cores capable of running parallel applications on separate threads. The threads communicate with each other via user-defined channels capable of transmitting data within the device. XMOS provides a C-based development environment using XC, which eliminates the need for custom tool kits associated with FPGA programming. The XC-2 has four cores and necessary hardware for Ethernet I/O.

Jarnot, Robert F.↗

Data flow modeling techniques

There have been a number of simulation packages developed for the purpose of designing, testing and validating computer systems, digital systems and software systems. Complex analytical tools based on Markov and semi-Markov processes have been designed to estimate the reliability and performance of simulated systems. Petri nets have received wide acceptance for modeling complex and highly parallel computers. In this research data flow models for computer systems are investigated. Data flow models can be used to simulate both software and hardware in a uniform manner. Data flow simulation techniques provide the computer systems designer with a CAD environment which enables highly parallel complex systems to be defined, evaluated at all levels and finally implemented in either hardware or software. Inherent in data flow concept is the hierarchical handling of complex systems. In this paper we will describe how data flow can be used to model computer system.

Kavi, K. M.↗

Operation of the Institute for Computer Applications in Science and Engineering

The ICASE research program is described in detail; it consists of four major categories: (1) efficient use of vector and parallel computers, with particular emphasis on the CDC STAR-100; (2) numerical analysis, with particular emphasis on the development and analysis of basic numerical algorithms; (3) analysis and planning of large-scale software systems; and (4) computational research in engineering and the natural sciences, with particular emphasis on fluid dynamics. The work in each of these areas is described in detail; other activities are discussed, a prognosis of future activities are included.

Source record↗

Accelerating Climate Simulations Through Hybrid Computing

Unconventional multi-core processors (e.g., IBM Cell B/E and NYIDIDA GPU) have emerged as accelerators in climate simulation. However, climate models typically run on parallel computers with conventional processors (e.g., Intel and AMD) using MPI. Connecting accelerators to this architecture efficiently and easily becomes a critical issue. When using MPI for connection, we identified two challenges: (1) identical MPI implementation is required in both systems, and; (2) existing MPI code must be modified to accommodate the accelerators. In response, we have extended and deployed IBM Dynamic Application Virtualization (DAV) in a hybrid computing prototype system (one blade with two Intel quad-core processors, two IBM QS22 Cell blades, connected with Infiniband), allowing for seamlessly offloading compute-intensive functions to remote, heterogeneous accelerators in a scalable, load-balanced manner. Currently, a climate solar radiation model running with multiple MPI processes has been offloaded to multiple Cell blades with approx.10% network overhead.

Zhou, Shujia↗

Adagio 2.9 user's guide.

Adagio is a Lagrangian, three-dimensional, implicit code for the analysis of solids and structures. It uses a multi-level iterative solver, which enables it to solve problems with large deformations, nonlinear material behavior, and contact. It also has a versatile library of continuum and structural elements, and an extensive library of material models. Adagio is written for parallel computing environments, and its solvers allow for scalable solutions of very large problems. Adagio uses the SIERRA Framework, which allows for coupling with other SIERRA mechanics codes. This document describes the functionality and input structure for Adagio.

97 MATHEMATICS AND COMPUTING↗

Grappolo

Community detection has become a fundamental operation in numerous graph-theoretic applications. It is used to reveal natural divisions that exist within real world networks without imposing prior size or cardinality constraints on the set of communities. Despite its potential for application, there is only limited support for community detection on large-scale parallel computers, largely owing to the irregular and inherently sequential nature of the underlying heuristics. In this paper, we present parallelization heuristics for fast community detection using the Louvain method as the serial template. The Louvain method is an iterative heuristic for modularity optimization. Originally developed by Blondel et al. in 2008, the method has become increasingly popular owing to its ability to detect high modularity community partitions in a fast and memory-efficient manner. However, the method is also inherently sequential, thereby limiting its scalability. Here, we observe certain key properties of this method that present challenges for its parallelization, and consequently propose heuristics that are designed to break the sequential barrier. For evaluation purposes, we implemented our heuristics using OpenMP multithreading, and tested them over real world graphs derived from multiple application domains (e.g., internet, citation, biological). Compared to the serial Louvain implementation, our parallel implementation is able to produce community outputs with a higher modularity for most of the inputs tested, in comparable number of iterations, while providing real speedups of up to 8x using 32 threads. In addition, our parallel implementation was able to exhibit weak scaling properties on up to 32 threads.

Halappanavar, Mahantesh↗

Piston geometry effects in a light-duty, swirl-supported diesel engine: Flow structure characterization

This work studied how in-cylinder flow structure is affected in a light-duty, swirl-supported diesel engine when equipped with three different piston geometries: the first two featuring a conventional re-entrant bowl, either with or without valve cut-outs on the piston surface and the third featuring a stepped-lip bowl. Particle image velocimetry experiments were conducted inside an optical engine to measure swirl vortex intensity and structure during the intake and compression strokes. A full computational model of the optical diesel engine was built using the FRESCO code, a recently developed object-oriented parallel computational fluid dynamics platform for engine simulations. The model was first validated against the measured swirl-plane velocity fields, and the simulation convergence for multiple cycles was assessed. Flow topology was studied by addressing bulk flow and turbulence quantities, including swirl structure, squish flux, plus geometric and operating parameters, such as the presence of valve cut-outs on the piston surface, compression ratio and engine speed. The results demonstrated that conventional re-entrant bowls have stronger flow separation at intake, hampering bowl swirl, but higher global swirl than for stepped-lip bowls thanks to a stronger and more axisymmetric squish mechanism and less tilted swirl. Stepped-lip bowls have larger inhomogeneities (tilt and axisymmetry) and higher turbulence levels, but also faster turbulence dissipation toward top dead center. They have weaker squish flux but larger squish inversion momentum as a result of the smaller inertia.

Perini, Federico↗

Algorithmic trends in computational fluid dynamics; The Institute for Computer Applications in Science and Engineering (ICASE)/LaRC Workshop, NASA Langley Research Center, Hampton, VA, US, Sep. 15-17, 1991

The purpose here is to assess the state of the art in the areas of numerical analysis that are particularly relevant to computational fluid dynamics (CFD), to identify promising new developments in various areas of numerical analysis that will impact CFD, and to establish a long-term perspective focusing on opportunities and needs. Overviews are given of discretization schemes, computational fluid dynamics, algorithmic trends in CFD for aerospace flow field calculations, simulation of compressible viscous flow, and massively parallel computation. Also discussed are accerelation methods, spectral and high-order methods, multi-resolution and subcell resolution schemes, and inherently multidimensional schemes.

Hussaini, M. Y.↗

MPI-IO: A Parallel File I/O Interface for MPI Version 0.3

Thanks to MPI [9], writing portable message passing parallel programs is almost a reality. One of the remaining problems is file I/0. Although parallel file systems support similar interfaces, the lack of a standard makes developing a truly portable program impossible. Further, the closest thing to a standard, the UNIX file interface, is ill-suited to parallel computing. Working together, IBM Research and NASA Ames have drafted MPI-I0, a proposal to address the portable parallel I/0 problem. In a nutshell, this proposal is based on the idea that I/0 can be modeled as message passing: writing to a file is like sending a message, and reading from a file is like receiving a message. MPI-IO intends to leverage the relatively wide acceptance of the MPI interface in order to create a similar I/0 interface. The above approach can be materialized in different ways. The current proposal represents the result of extensive discussions (and arguments), but is by no means finished. Many changes can be expected as additional participants join the effort to define an interface for portable I/0. This document is organized as follows. The remainder of this section includes a discussion of some issues that have shaped the style of the interface. Section 2 presents an overview of MPI-IO as it is currently defined. It specifies what the interface currently supports and states what would need to be added to the current proposal to make the interface more complete and robust. The next seven sections contain the interface definition itself. Section 3 presents definitions and conventions. Section 4 contains functions for file control, most notably open. Section 5 includes functions for independent I/O, both blocking and nonblocking. Section 6 includes functions for collective I/O, both blocking and nonblocking. Section 7 presents functions to support system-maintained file pointers, and shared file pointers. Section 8 presents constructors that can be used to define useful filetypes (the role of filetypes is explained in Section 2 below). Section 9 presents how the error handling mechanism of MPI is supported by the MPI-IO interface. All this is followed by a set of appendices, which contain information about issues that have not been totally resolved yet, and about design considerations. The reader can find there the motivation behind some of our design choices. More information on this would definitely be welcome and will be included in a further release of this document. The first appendix contains a description of MPI-I0's 'hints' structure which is used when opening a file. Appendix B is a discussion of various issues in the support for file pointers. Appendix C explains what we mean in talking about atomic access. Appendix D provides detailed examples of filetype constructors, and Appendix E contains a collection of arguments for and against various design decisions.

Corbett, Peter↗

NCC: A Physics-Based Design and Analysis Tool for Combustion Systems

The National Combustion Code (NCC) is an integrated system of computer codes for physics-based design and analysis of combustion systems. It uses unstructured meshes and runs on parallel computing platforms. The NCC is composed of a set of distinct yet closely related modules. They are: (1) a gaseous flow module solving 3-D Navier-Stokes equations; (2) a turbulence module containing the non-linear k-epsilon models; (3) a chemistry module using either the conventional reduced kinetics approach of solving species equations or the Intrinsic Low Dimensional Manifold (ILDM) kinetics approach of table looking up in conjunction with solving the equations of the progressive variables; (4) a turbulence-chemistry interaction module including the option of solving the joint probability density function (PDF) for species and enthalpy; and (5) a spray module for solving the liquid phase equations. In early 1995, an industry-government team was formed to develop the NCC. In July 1998, the baseline beta version was completed and presented in two NCC sessions at the 34th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit, July 1998. An overview of this baseline beta version was presented at the NASA HPCCP/CAS Workshop 98, August 1998. Since then, the effort has been focused on the streamlining, validation, and enhancement of the th baseline beta version. The progress is presented in two NCC sessions at the AIAA 38 Aerospace Sciences Meeting & Exhibit, January 2000. At this NASA HPCCP/CAS Workshop 2000, an overview of the NCC papers presented at the AIAA 38 th Aerospace Sciences Meeting & Exhibit is presented, with emphasis on the reduction of analysis time of simulating the (gaseous) reacting flows in full combustors. In addition, results of NCC simulation of a modern turbofan combustor will also be reported.

Liu, Nan-Suey↗

Parallel Heuristics for Scalable Community Detection

Community detection has become a fundamental operation in numerous graph-theoretic applications. It is used to reveal natural divisions that exist within real world networks without imposing prior size or cardinality constraints on the set of communities. Despite its potential for application, there is only limited support for community detection on large-scale parallel computers, largely owing to the irregular and inherently sequential nature of the underlying heuristics. In this paper, we present parallelization heuristics for fast community detection using the Louvain method as the serial template. The Louvain method is an iterative heuristic for modularity optimization. Originally developed by Blondel et al. in 2008, the method has become increasingly popular owing to its ability to detect high modularity community partitions in a fast and memory-efficient manner. However, the method is also inherently sequential, thereby limiting its scalability to problems that can be solved on desktops. Here, we observe certain key properties of this method that present challenges for its parallelization, and consequently propose multiple heuristics that are designed to break the sequential barrier. Our heuristics are agnostic to the underlying parallel architecture. For evaluation purposes, we implemented our heuristics on shared memory (OpenMP) and distributed memory (MapReduce-MPI) machines, and tested them over real world graphs derived from multiple application domains (internet, biological, natural language processing). Experimental results demonstrate the ability of our heuristics to converge to high modularity solutions comparable to those output by the serial algorithm in nearly the same number of iterations, while also drastically reducing time to solution.

clustering, graph algorithms↗

Maestro: Data Orchestation and Tuning for OpenCL Devices

Abstract. As heterogeneous computing platforms become more prevalent, the programmer must account for complex memory hierarchies in addition to the diculties of parallel programming. OpenCL is an open standard for parallel computing that helps alleviate this diculty by providing a portable set of abstractions for device memory hierarchies. However, OpenCL requires that the programmer explicitly controls data transfer and device synchronization, two tedious and error-prone tasks. This paper introduces Maestro, an open source library for data orchestration on OpenCL devices. Maestro provides automatic data transfer, task decomposition across multiple devices, and autotuning of dynamic execution parameters for some types of problems.

99 GENERAL AND MISCELLANEOUS↗

Parallel Agent-Based Simulations on Clusters of GPUs and Multi-Core Processors

An effective latency-hiding mechanism is presented in the parallelization of agent-based model simulations (ABMS) with millions of agents. The mechanism is designed to accommodate the hierarchical organization as well as heterogeneity of current state-of-the-art parallel computing platforms. We use it to explore the computation vs. communication trade-off continuum available with the deep computational and memory hierarchies of extant platforms and present a novel analytical model of the tradeoff. We describe our implementation and report preliminary performance results on two distinct parallel platforms suitable for ABMS: CUDA threads on multiple, networked graphical processing units (GPUs), and pthreads on multi-core processors. Message Passing Interface (MPI) is used for inter-GPU as well as inter-socket communication on a cluster of multiple GPUs and multi-core processors. Results indicate the benefits of our latency-hiding scheme, delivering as much as over 100-fold improvement in runtime for certain benchmark ABMS application scenarios with several million agents. This speed improvement is obtained on our system that is already two to three orders of magnitude faster on one GPU than an equivalent CPU-based execution in a popular simulator in Java. Thus, the overall execution of our current work is over four orders of magnitude faster when executed on multiple GPUs.

97 MATHEMATICS AND COMPUTING↗

Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)

The development of modern processors exhibits two trends that complicate the optimizations of modern software. The first is the increasing sensitivity of processors' throughput to irregularities in computation. With more processors produced through a massive integration of simple cores, future systems will increasingly favor regular data-level parallel computations, but deviate from the needs of applications with complex patterns. Some evidences are already shown on Graphic Processing Units (GPU): Irregular data accesses (e.g., indirect references A[D[i]]) and conditional branches are limiting many GPU applications' performance at a level an order of magnitude lower than the peak of GPU. The second hardware trend is the growing gap between memory bandwidth and the aggregate speed—that is, the sum of all cores' computing power—of a Chip Multiprocessor (CMP). Despite the capped growth of the peak CPU speed, the aggregate speed of a CMP keeps increasing as more cores get into a single chip. It is expected that by 2018, node concurrency in an exascale system will increase by hundreds of times, whereas, memory bandwidth will expand by only 10 to 20 times. Consequently, data movement and storage is expected to consume more than 70% of the total system power. Bridging this gap is difficult; the complexities of modern CMP memory hierarchy make it even harder: Data cache becomes shared among computing units, and the sharing is often non-uniform—whether two computing units share a cache depends on their proximity and the level of the cache. On the recent IBM Power7 architecture, for instance, four hardware contexts (or SMT threads) in a core share the entire memory hierarchy, all cores in one chip share an on-chip L3 cache, and cores across chips share L3 and main memory through off-chip connections. These two trends complicate the translation of computing power into performance, especially for a program with either intensive data accesses or complex patterns in data accesses or control flow paths. Unfortunately, both attributes present and will persist in a class of important applications. For instance, many scientific simulations deal with a large volume of data. And meanwhile, as most real-world processes are non-uniform and evolving (e.g., the evolution of a galaxy or the process of a drug injection), both the computations and data accesses of these programs tend to be irregular and dynamically changing. Currently, the lack of support to these applications on modern CMP severely limits their performance. On GPU, as our recent study shows and other studies echo, performance enhancement of a factor of integers is possible when memory accesses or control flows are streamlined for a set of GPU applications. On multicore CPU, our studies show that traditional locality enhancement, for being oblivious to the new features of multicore memory hierarchy, may even cause large slowdown to data-intensive dynamic applications. The severity of the issues is expected to worsen as the two hardware trends continue. Some recent studies try to match software with the trends, but in a limited scope or manner. For irregularities on GPU, most studies focus on irregularities analyzable through static analysis (e.g., data accesses in regular loops). Dynamic irregularities are harder to address because the needed analysis and transformations typically have to happen at run time. Some other research resorts to hardware extensions, an actual adoption of which is unclear for the entailed space cost and complexity. For data locality, recent years have seen some exploitations of the new memory hierarchy on multicore for performance, but most of them are on process or thread scheduling, rather than program transformations. Our recent study reveals that program-level transformations may magnify the scheduling benefits by a factor of seven, concluding that program-level transformation should play a central role for data locality enhancement on modern CMP. But research in this direction has been sparse, and most have focused on data layout or cache performance modeling, rather than program transformations to match with the new memory hierarchy features. Overall, it is still an open question how to bridge the gap between dynamic computations and the two prominent properties of modern processors. The goal of this project is to develop a set of techniques and software tools to enhance the matching between memory accesses in dynamic simulations and the prominent features of modern and future CMP, alleviating the memory performance issues for petascale and exascale computing. This report summarizes the discoveries and products produced throughout this project. It includes free launch, a new software approach to overcoming the shortcomings of both methods; coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution; algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Supporting dynamic parallelism is important for GPU to benefit a broad range of applications. There are currently two fundamental ways for programs to exploit dynamic parallelism on GPU: a software-based approach with software-managed worklists, and a hardware-based approach through dynamic subkernel launches. Neither is satisfactory. The former is complicated to program and is often subject to some load imbalance; the latter suffers large runtime overhead. In this work, we propose free launch, a new software approach to overcoming the shortcomings of both methods. It allows programmers to use subkernel launches to express dynamic parallelism. It employs a novel compiler-based code transformation named subkernel launch removal to replace the subkernel launches with the reuse of parent threads. Coupled with an adaptive task assignment mechanism, the transformation reassigns the tasks in the subkernels to the parent threads with a good load balance. The technique requires no hardware extensions, immediately deployable on existing GPUs. It keeps the programming convenience of the subkernel launch-based approach while avoiding its large runtime overhead. Meanwhile, its superior load balancing makes it outperform manual worklist-based techniques by 3X on average. The work was published at the 48th Annual IEEE/ACM International Symposium on Microarchitecture (Micro'2015). A Graphic Processing Unit (GPU) system is typically equipped with many types of memory (e.g., global, constant, texture, shared, cache). Data placement determines what data are placed on which type of memory, essential for GPU memory performance. Prior optimizations of data placement always require a single view of a data object on memory, which limits the optimization effectiveness. In this work, we propose coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution. We demonstrate that under certain conditions, the multiple views can remain incoherent while facilitating enhanced data placement. We present a theorem and some compiler support to ensure the soundness of the usage of coherence-free multiview. We further develop reference-discerning data placement, a new way to enhance data placements on GPU. It enables more flexible data placements by using coherence-free multiview to leverage the slack in coherence requirement of some GPU programs. Experiments on three types of GPU systems show that, with less than 200KB space cost, the new data placement technique can provide a 1.6X average (up to 4.27X) speedup. In addition, we have examined algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Computing distances among data points is an essential part of many important algorithms in data analytics, graph analysis, and other domains. In each of these domains, developers have spent significant manual effort optimizing algorithms, often through novel applications of the triangle equality, in order to minimize the number of distance computations in the algorithms. In this work, we observe that many algorithms across these domains can be generalized as an instance of a generic distance-related abstraction. Based on this abstraction, we derive seven principles for correctly applying the triangular inequality to optimize distance-related algorithms. Guided by the findings, we develop Triangular OPtimizer (TOP), the first software framework that is able to automatically produce optimized algorithms that either matches or outperforms manually designed algorithms for solving distance-related problems. TOP achieves up to 237x speedups and 2.5X on average. The work has been published at the 32nd International Conference on Machine Learning (ICML'15) and the 41st International Conference on Very Large Data Bases (VLDB'15).

42 ENGINEERING↗