SEARCH · Engineering Papers

Results for “parallel performance”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 289 records · Page 16

All sources NASA NTRS DOE OSTI

NASA NTRS1989

NETRA: A parallel architecture for integrated vision systems. 1: Architecture and organization

Computer vision is regarded as one of the most complex and computationally intensive problems. An integrated vision system (IVS) is considered to be a system that uses vision algorithms from all levels of processing for a high level application (such as object recognition). A model of computation is presented for parallel processing for an IVS. Using the model, desired features and capabilities of a parallel architecture suitable for IVSs are derived. Then a multiprocessor architecture (called NETRA) is presented. This architecture is highly flexible without the use of complex interconnection schemes. The topology of NETRA is recursively defined and hence is easily scalable from small to large systems. Homogeneity of NETRA permits fault tolerance and graceful degradation under faults. It is a recursively defined tree-type hierarchical architecture where each of the leaf nodes consists of a cluster of processors connected with a programmable crossbar with selective broadcast capability to provide for desired flexibility. A qualitative evaluation of NETRA is presented. Then general schemes are described to map parallel algorithms onto NETRA. Algorithms are classified according to their communication requirements for parallel processing. An extensive analysis of inter-cluster communication strategies in NETRA is presented, and parameters affecting performance of parallel algorithms when mapped on NETRA are discussed. Finally, a methodology to evaluate performance of algorithms on NETRA is described.

Choudhary, Alok N.↗

DOE OSTI2022

Stable parallel training of Wasserstein conditional generative adversarial neural networks

In this work, we propose a stable, parallel approach to train Wasserstein conditional generative adversarial neural networks (W-CGANs) under the constraint of a fixed computational budget. Differently from previous distributed GANs training techniques, our approach avoids inter-process communications, reduces the risk of mode collapse and enhances scalability by using multiple generators, each one of them concurrently trained on a single data label. The use of the Wasserstein metric also reduces the risk of cycling by stabilizing the training of each generator. We illustrate the approach on the CIFAR10, CIFAR100, and ImageNet1k datasets, three standard benchmark image datasets, maintaining the original resolution of the images for each dataset. Performance is assessed in terms of scalability and final accuracy within a limited fixed computational time and computational resources. To measure accuracy, we use the inception score, the Fréchet inception distance, and image quality. An improvement in inception score and Fréchet inception distance is shown in comparison to previous results obtained by performing the parallel approach on deep convolutional conditional generative adversarial neural networks as well as an improvement of image quality of the new images created by the GANs approach. Weak scaling is attained on both datasets using up to 2000 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1992

Comparison of optics and electronics for the calculation of matrix-vector products

Optical processors are attractive because of their ability to perform massively parallel operations such as matrix vector products. The inherently analog nature of optical calculations requires that optical processors be based on analog computations. While the speed at which such analog operations can be performed as well as the natural parallelism of optical systems are great advantages of optical processors, the analog representation of values severely limits the achievable accuracy. Furthermore, optical processors are limited by the need to convert information to and from the intensity of light. Digitization can be used to increase the accuracy of optical matrix-vector processors, but causes a severe reduction in speed. This paper compares the throughput and power requirements of optical and electronic processors, showing that optical matrix-vector processors can provide a greater number of operations/Watt than conventional electronics.

Gary, C. K.↗

NASA NTRS1992

Turbomachinery CFD on parallel computers

The role of multistage turbomachinery simulation in the development of propulsion system models is discussed. Particularly, the need for simulations with higher fidelity and faster turnaround time is highlighted. It is shown how such fast simulations can be used in engineering-oriented environments. The use of parallel processing to achieve the required turnaround times is discussed. Current work by several researchers in this area is summarized. Parallel turbomachinery CFD research at the NASA Lewis Research Center is then highlighted. These efforts are focused on implementing the average-passage turbomachinery model on MIMD, distributed memory parallel computers. Performance results are given for inviscid, single blade row and viscous, multistage applications on several parallel computers, including networked workstations.

Blech, Richard A.↗

NASA NTRS1993

The development of a scalable parallel 3-D CFD algorithm for turbomachinery

Two algorithms capable of computing a transonic 3-D inviscid flow field about rotating machines are considered for parallel implementation. During the study of these algorithms, a significant new method of measuring the performance of parallel algorithms is developed. The theory that supports this new method creates an empirical definition of scalable parallel algorithms that is used to produce quantifiable evidence that a scalable parallel application was developed. The implementation of the parallel application and an automated domain decomposition tool are also discussed.

Luke, Edward Allen↗

NASA NTRS1992

Turbomachinery CFD on parallel computers

Blech, R. A.↗

NASA NTRS2012

Airborne Precision Spacing for Dependent Parallel Operations Interface Study

This paper describes a usability study of proposed cockpit interfaces to support Airborne Precision Spacing (APS) operations for aircraft performing dependent parallel approaches (DPA). NASA has proposed an airborne system called Pair Dependent Speed (PDS) which uses their Airborne Spacing for Terminal Arrival Routes (ASTAR) algorithm to manage spacing intervals. Interface elements were designed to facilitate the input of APS-DPA spacing parameters to ASTAR, and to convey PDS system information to the crew deemed necessary and/or helpful to conduct the operation, including: target speed, guidance mode, target aircraft depiction, and spacing trend indication. In the study, subject pilots observed recorded simulations using the proposed interface elements in which the ownship managed assigned spacing intervals from two other arriving aircraft. Simulations were recorded using the Aircraft Simulation for Traffic Operations Research (ASTOR) platform, a medium-fidelity simulator based on a modern Boeing commercial glass cockpit. Various combinations of the interface elements were presented to subject pilots, and feedback was collected via structured questionnaires. The results of subject pilot evaluations show that the proposed design elements were acceptable, and that preferable combinations exist within this set of elements. The results also point to potential improvements to be considered for implementation in future experiments.

Volk, Paul M.↗

NASA NTRS1993

A fast algorithm for parallel computation of multibody dynamics on MIMD parallel architectures

In this paper the implementation of a parallel O(LogN) algorithm for computation of rigid multibody dynamics on a Hypercube MIMD parallel architecture is presented. To our knowledge, this is the first algorithm that achieves the time lower bound of O(LogN) by using an optimal number of O(N) processors. However, in addition to its theoretical significance, the algorithm is also highly efficient for practical implementation on commercially available MIMD parallel architectures due to its highly coarse grain size and simple communication and synchronization requirements. We present a multilevel parallel computation strategy for implementation of the algorithm on a Hypercube. This strategy allows the exploitation of parallelism at several computational levels as well as maximum overlapping of computation and communication to increase the performance of parallel computation.

Fijany, Amir↗

NASA NTRS1988

Performance of a class of parallel spread-spectrum code acquisition schemes in the presence of data modulation

An acquisition procedure for data-modulated direct-sequence spread-spectrum systems is investigated. The correlation time is partitioned into subintervals, and the integration results in these subintervals are noncoherently combined for detection. The tradeoff between noncoherent combining loss and data modulation degradation guides the optimum choice of the number of subintervals. Two forms of data modulation are considered, namely, the alternate-data and random-data cases. The parallel acquisition schemes discussed allow multiple code-phase offsets to be examined at each test. The circular state diagram approach is used to analyze the performances of these schemes. The theory presented is valid for a class of such parallel schemes.

Cheng, Unjeng↗

NASA NTRS1996

Performance Analysis and Optimization on the UCLA Parallel Atmospheric General Circulation Model Code

An analysis is presented of several factors influencing the performance of a parallel implementation of the UCLA atmospheric general circulation model (AGCM) on massively parallel computer systems. Several modificaitons to the original parallel AGCM code aimed at improving its numerical efficiency, interprocessor communication cost, load-balance and issues affecting single-node code performance are discussed.

atmospheric study optimization strategies parallel↗

DOE OSTI2021

Scalable balanced training of conditional generative adversarial neural networks on image data

Here, we propose a distributed approach to train deep convolutional generative adversarial neural network (DC-CGANs) models. Our method reduces the imbalance between generator and discriminator by partitioning the training data according to data labels, and enhances scalability by performing a parallel training where multiple generators are concurrently trained, each one of them focusing on a single data label. Performance is assessed in terms of inception score, Fréchet inception distance, and image quality on MNIST, CIFAR10, CIFAR100, and ImageNet1k datasets, showing a significant improvement in comparison to state-of-the-art techniques to training DC-CGANs. Weak scaling is attained on all the four datasets using up to 1000 processes and 2000 NVIDIA V100 GPUs on the OLCF supercomputer Summit.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS1986

Application of data flow concepts to a multigrid solver for the Euler equations

In this study a multigrid solver for Euler equations (FLO52R) was examined to determine its performance potential on a hypothetical computer using a data flow architecture. The proposed computer would require massive parallelism to realize its design performance. On the other hand this parallelism would be more easily realized than with a conventional vector processor such as the Cray-1S. Several changes to the proposed design substantially alleviated most of the remaining bottlenecks to parallel processing. Other changes allowed clearer definition of memory access and disk I/O. Finally, a portion of the algorithm was rewritten to improve parallel performance. With these changes, performance levels approaching that of a Cray-1S may be possible for a computer costing far less. Estimates are given for overall speed, memory, and network bandwidth, and for instruction memory requirements.

Merriam, M. L.↗

NASA NTRS1988

BLITZEN - A highly integrated massively parallel machine

The architecture and VLSI design of a new massively parallel processing array chip are described. The BLITZEN processing element array chip, which contains 1.1 million transistors, serves as the basis for a highly integrated, miniaturized, high-performance, massively parallel machine that is currently under development. Each processing element has 1K bits of static RAM and performs bit-serial processing with functional elements for arithmetic, logic, and shifting.

Blevins, D. W.↗

NASA NTRS1988

BLITZEN: A highly integrated massively parallel machine

Blevins, D. W.↗

NASA NTRSDate not supplied

Parametric Modeling and Mission Performance Analysis of a True Parallel Hybrid Turboprop Aircraft for Freighter Operations

Hybrid-electric propulsion systems for short-haul, cargo carrying aircraft have emerged as promising solutions for an environmentally sustainable future for commercial aviation. The novel propulsion architecture presents significant complexity and requires the development of new methodologies to account for the unique coupling and integration between critical design variables. This paper presents a comprehensive study on the modeling, performance assessment, and design space exploration of a C-130H freighter aircraft retrofitted with a parallel hybrid electric powertrain. Details on the development of parametric models for both the baseline and true parallel hybrid (TPH) aircraft and the integrated electrified aircraft propulsion (EAP) system sizing approach are presented along with detailed performance analyses of payload/range capabilities for short-haul cargo missions. For 2030, 2040, and 2050 EAP technology levels, the TPH C-130H configuration with a ~2.12 MW class EAP system has a range capability of 485-1,028 nautical miles and block fuel savings of 27-44%.

efficiency↗

DOE OSTI2023

ORNL/grit

Grit is a software library designed specifically for conducting particle-based Lagrangian simulations with CPU/GPUperformance portability. This library allows researchers and engineers to perform a wide range of simulations, including multiphase flow, and more. Grit employs Message Passing Interface (MPI) for distributed memory parallelism and Kokkos programming model for on-node shared memory parallelism with performance portability across different architectures of GPUs and multi-core/manycore CPUs.

Ge, Wenjun [Oak Ridge National Laboratory (ORNL), ↗

NASA NTRS1992

Scalable parallel communications

Coarse-grain parallelism in networking (that is, the use of multiple protocol processors running replicated software sending over several physical channels) can be used to provide gigabit communications for a single application. Since parallel network performance is highly dependent on real issues such as hardware properties (e.g., memory speeds and cache hit rates), operating system overhead (e.g., interrupt handling), and protocol performance (e.g., effect of timeouts), we have performed detailed simulations studies of both a bus-based multiprocessor workstation node (based on the Sun Galaxy MP multiprocessor) and a distributed-memory parallel computer node (based on the Touchstone DELTA) to evaluate the behavior of coarse-grain parallelism. Our results indicate: (1) coarse-grain parallelism can deliver multiple 100 Mbps with currently available hardware platforms and existing networking protocols (such as Transmission Control Protocol/Internet Protocol (TCP/IP) and parallel Fiber Distributed Data Interface (FDDI) rings); (2) scale-up is near linear in n, the number of protocol processors, and channels (for small n and up to a few hundred Mbps); and (3) since these results are based on existing hardware without specialized devices (except perhaps for some simple modifications of the FDDI boards), this is a low cost solution to providing multiple 100 Mbps on current machines. In addition, from both the performance analysis and the properties of these architectures, we conclude: (1) multiple processors providing identical services and the use of space division multiplexing for the physical channels can provide better reliability than monolithic approaches (it also provides graceful degradation and low-cost load balancing); (2) coarse-grain parallelism supports running several transport protocols in parallel to provide different types of service (for example, one TCP handles small messages for many users, other TCP's running in parallel provide high bandwidth service to a single application); and (3) coarse grain parallelism will be able to incorporate many future improvements from related work (e.g., reduced data movement, fast TCP, fine-grain parallelism) also with near linear speed-ups.

Maly, K.↗

NASA NTRSDate not supplied

Performance Optimization Methods for a Memory-Bound, Unstructured-Grid CFD Application on Massively Parallel GPU Platforms

Computational performance of the FUN3D unstructured-grid computational fluid dynamics (CFD) application on massively parallel GPU environments is memory-bound and highly dependent upon efficient reads from and atomic updates to the irregular cell-, edge-, and node-based data structures. In this talk, we present recent efforts into optimizing select performance-critical kernels on NVIDIA Tesla V100 and A100 GPUs and AMD CDNA MI100 GPUs. A novel use of L2 cache residency controls and asynchronous loads into on-chip shared memory are explored on the A100 GPU for the sparse iterative solver, which is dominated by mixed-precision, sparse matrix vector multiplication. Demonstrations show that these methods improve global memory bandwidth utilization by 13.5% on the A100 GPU. Several techniques are also presented that use registers and/or shared memory to facilitate array transposition and aggregation which combine to reduce the frequency and increase the cache efficiency of floating-point atomic updates to the irregular data structures. These methods are demonstrated to improve the kernel throughput by nearly 500% on select kernels on the AMD MI100 over atomic updates directly to global memory. Overall, both V100 and A100 GPUs outperformed the MI100 GPU on kernels dominated by double-precision atomic updates; however, the techniques demonstrated here reduced the performance gap and improved the MI100 performance.

GPU CPU unstructured CFD memory↗