Search NASA⌕ Search

SEARCH · Search NASA

Results for “asynchronous processing”

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 217 records · Page 12

State Estimation for Distribution Networks with Asynchronous Sensors Using Stochastic Descent: Preprint

This paper investigates the problem of state estimation for distribution networks with asynchronous sensors comprising of a mix of smart meters and phasor measurement units (PMUs) with multiple sampling and reporting rates. We consider two independent scenarios of state estimation and tracking, with either voltages or currents as states. With these two sets, we investigate estimation under (a) full data, assuming all measurements are available and (b) limited data, where an online algorithmic approach is adopted to estimate the possibly time-varying states by processing measurements as and when available. The proposed algorithm, inspired by the classical Stochastic Gradient Descent (SGD) approach updates the states based on the previous estimate and the newly available measurements. Finally, we demonstrate the estimation and tracking efficacy through numerical simulations on the IEEE-37 test network, while also highlighting how estimation with currents as states leads to faster convergence.

asynchronous sensors↗

High-order asynchrony-tolerant finite difference schemes for partial differential equations

Synchronizations of processing elements (PEs) in massively parallel simulations, which arise due to communication or load imbalances between PEs, significantly affect the scalability of scientific applications. We have recently proposed a method based on finite-difference schemes to solve partial differential equations in an asynchronous fashion – synchronization between PEs is relaxed at a mathematical level. While standard schemes can maintain their stability in the presence of asynchrony, their accuracy is drastically affected. In this work, we present a general methodology to derive asynchrony-tolerant (AT) finite difference schemes of arbitrary order of accuracy, which can maintain their accuracy when synchronizations are relaxed. We show that there are several choices available in selecting a stencil to derive these schemes and discuss their effect on numerical and computational performance. We provide a simple classification of schemes based on the stencil and derive schemes that are representative of different classes. Their numerical error is rigorously analyzed within a statistical framework to obtain the overall accuracy of the solution. Results from numerical experiments are used to validate the performance of the schemes.

42 ENGINEERING↗

Hybrid additive manufacturing of AISI 316L via asynchronous powder and hot-wire laser directed energy deposition

Hybrid Additive Manufacturing (AM) offers a way to leverage the advantages of different AM technologies, enabling the efficient production of sizeable parts without compromising material properties or geometric complexity capabilities. This study presents an asynchronous hybrid Directed Energy Deposition (DED) strategy employing laser powder DED and laser hot-wire DED. AISI 316L parts comprising multiple powder and wire segments were fabricated with optional machining on AISI 316L substrates to investigate how quality is impacted by (i) alternative process sequences (laser powder DED followed by laser hot-wire DED and vice versa), (ii) machined vs. as-printed interfacial conditions, and (iii) material deposition on top vs. alongside previously built segments. Optical microscopy, X-ray computed tomography, and Vickers hardness were used to characterize the morphology and microstructure of the parts, localized porosity and lack of fusion defects, bulk density, and mechanical properties. Interfacial machining was necessary for dimensional control but promoted lack of fusion voids, resulting in a 99.71 ± 0.01% dense part. As-printed interfaces resulted in a denser part (99.82 ± 0.02%) at the expense of dimensional accuracy. The hardness of the parts with as-printed and machined interfaces was 196 ± 0.37 HV and 192 ± 0.40 HV, respectively, compared to 156 ± 1.4 HV for the substrate. Depositing powder alongside or on top of wire sections resulted in interfaces with a hardness of 217 ± 2.2 HV, compared to 185 ± 3.4 HV for the wire-powder interfaces.

36 MATERIALS SCIENCE↗

Multi-star processing and gyro filtering for the video inertial pointing system

The video inertial pointing (VIP) system is being developed to satisfy the acquisition and pointing requirements of astronomical telescopes. The VIP system uses a single video sensor to provide star position information that can be used to generate three-axis pointing error signals (multi-star processing) and for input to a cathode ray tube (CRT) display of the star field. The pointing error signals are used to update the telescope's gyro stabilization system (gyro filtering). The CRT display facilitates target acquisition and positioning of the telescope by a remote operator. Linearized small angle equations are used for the multistar processing and a consideration of error performance and singularities lead to star pair location restrictions and equation selection criteria. A discrete steady-state Kalman filter which uses the integration of the gyros is developed and analyzed. The filter includes unit time delays representing asynchronous operations of the VIP microprocessor and video sensor. A digital simulation of a typical gyro stabilized gimbal is developed and used to validate the approach to the gyro filtering.

Murphy, J. P.↗

Asynchronicity in opposed-piston RCMs: Does it matter?

Rapid Compression Machines (RCMs) are widely utilized to study combustion phenomena at engine-relevant conditions, and significant efforts are typically made to create a quiescent environment, particularly for investigations of autoignition chemistry. Opposed-piston configurations can be advantageous due to shorter compression times and reduced surface area to volume ratios. Each side must be actuated simultaneously, but this can be challenging in practice. These devices, like most RCMs, utilize hydraulics for actuation, speed control and arrestation of the piston at the end of the stroke; there is no mechanical control or linkage of the two piston trajectories. To quantify the magnitudes and effects of piston asynchronous behavior, this work employs both detailed experimental measurements and, for the first time, high-fidelity, Direct Numerical Simulation (DNS). The boundary conditions are carefully considered applying insight from high-resolution linear variable differential transformer (LVDT) measurements of the piston trajectory and a zero-dimensional kinematics model of the piston-shaft assembly. Sufficient resolution in the piston crevice region is used. The complicated fluid dynamical behavior that can evolve during piston compression and the ensuing delay processes due to offset timings from t offset = 0-10 ms is elucidated. It is found that near t offset = 6 ms and beyond, the boundary layer on the face of the first-seating piston can be sufficiently perturbed, due initially to reemergence of gas from the crevice of the firstseating piston, so that the adiabatic core can become degraded at long ignition delay times. Substantial mixing of colder gas into the interior of the reaction chamber can alter the measurements, similar to effects previously observed for improper piston crevice configuration. In conclusion, experimental techniques to mitigate asynchronous behavior are discussed and demonstrated.

33 ADVANCED PROPULSION SYSTEMS↗

UPC++ Programmer’s Guide (v1.0-2019.3.0)

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

UPC++ v1.0 Programmer’s Guide (Revision 2019.9.0)

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

UPC++ v1.0 Programmer’s Guide, Revision 2020.10.0

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

UPC++ Programmer’s Guide, v1.0-2018.9.0

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

UPC++ v1.0 Programmer’s Guide, Revision 2020.3.0

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗

Development of a high capacity bubble domain memory element and related epitaxial garnet materials for application in spacecraft data recorders. Item 1: Development of a high capacity memory element

Several versions of the 100K bit chip, which is configured as a single serial loop, were designed, fabricated and evaluated. Design and process modifications were introduced into each succeeding version to increase device performance and yield. At an intrinsic field rate of 150 KHz the final design operates from -10 C to +60 C with typical bias margins of 12 and 8 percent, respectively, for continuous operation. Asynchronous operation with first bit detection on start-up produces essentially the same margins over the temperature range. Cost projections made from fabrication yield runs on the 100K bit devices indicate that the memory element cost will be less than 10 millicents/bit in volume production.

Besser, P. J.↗

Bi-directional four quadrant (BDQ4) power converter development

The feasibility for implementation of a concept for direct ac/dc multikilowatt power conversion with bidirectional transfer of energy was investigated. A 10 kHz current carrier was derived directly from a common 60 Hz three phase power system. This carrier was modulated to remove the 360 Hz ripple, inherent in the three phase power supply and then demodulated and processed by a high frequency filter. The resulting dc power was then supplied to a load. The process was implemented without the use of low frequency transformers and filters. This power conversion processes was reversible and can operate in the four quadrants as viewed from any of the two of the converter's ports. Areas of application include: power systems on air and spacecraft; terrestrial traction; integration of solar and wind powered systems with utility networks; HVDC; asynchronous coupling of polyphase networks; heat treatment; industrial machine drives; and power supplies for any use including instrumentation.

Schwarz, F. C.↗

Asynchronous Message Passing in the JPL Flight System Testbed

The flight mission simulation software in the Jet Propulsion Laboratory's Flight System Testbed (FST) is a heterogeneous, distributed system that is built on an interprocess communication model of asynchronous message passing rather than remote procedure calls (RPCs). The reasoning behind this design decision is discussed; the mechanism used to implement it (.

message passing RPC interprocess communication dis↗

A robust low-rate coding scheme for packet video

Due to the rapidly evolving field of image processing and networking, video information promises to be an important part of telecommunication systems. Although up to now video transmission has been transported mainly over circuit-switched networks, it is likely that packet-switched networks will dominate the communication world in the near future. Asynchronous transfer mode (ATM) techniques in broadband-ISDN can provide a flexible, independent and high performance environment for video communication. For this paper, the network simulator was used only as a channel in this simulation. Mixture blocking coding with progressive transmission (MBCPT) has been investigated for use over packet networks and has been found to provide high compression rate with good visual performance, robustness to packet loss, tractable integration with network mechanics and simplicity in parallel implementation.

Chen, Y. C.↗

ARENA: Asynchronous Reconfigurable Accelerator Ring to Enable Data-Centric Parallel Computing

The next generation HPC and data centers are likely to be reconfigurable and data-centric due to the trend of hardware specialization and the emergence of data-driven applications. In this work, we propose ARENA – an asynchronous reconfigurable accelerator ring architecture as a potential scenario on how the future HPC and data centers will be like. Despite using the coarse-grained reconfigurable arrays (CGRAs) as the substrate platform, our key contribution is not only the CGRA-cluster design itself, but also the ensemble of a new architecture and programming model that enables asynchronous tasking across a cluster of reconfigurable nodes, so as to bring specialized computation to the data rather than the reverse. We presume distributed data storage without asserting any prior knowledge on the data distribution. Hardware specialization occurs at runtime when a task finds the majority of data it requires are available at the present node. In other words, we dynamically generate specialized CGRA accelerators where the data reside. The asynchronous tasking for bringing computation to data is achieved by circulating the task token, which describes the dataflow graphs to be executed for a task, among the CGRA cluster connected by a fast ring network. Evaluations on a set of HPC and data-driven applications across different domains show that ARENA can provide better parallel scalability with reduced data movement (53.9 percent). Compared with contemporary compute-centric parallel models, ARENA can bring on average 4.37× speedup. The synthesized CGRAs and their task-dispatchers only occupy 2.93mm 2 chip area under 45nm process technology and can run at 800MHz with on average 759.8mW power consumption. ARENA also supports the concurrent execution of multi-applications, offering ideal architectural support for future high-performance parallel computing and data analytics systems.

97 MATHEMATICS AND COMPUTING↗

Asynchronous and Load-Balanced Union-Find for Distributed and Parallel Scientific Data Visualization and Analysis

We present a novel distributed union-find algorithm that features asynchronous parallelism and k-d tree based load balancing for scalable visualization and analysis of scientific data. Applications of union-find include level set extraction and critical point tracking, but distributed union-find can suffer from high synchronization costs and imbalanced workloads across parallel processes. In this study, we prove that global synchronizations in existing distributed union-find can be eliminated without changing final results, allowing overlapped communications and computations for scalable processing. We also use a k-d tree decomposition to redistribute inputs, in order to improve workload balancing. We benchmark the scalability of our algorithm with up to 1,024 processes using both synthetic and application data. Here, we demonstrate the use of our algorithm in critical point tracking and super-level set extraction with high-speed imaging experiments and fusion plasma simulations, respectively.

97 MATHEMATICS AND COMPUTING↗

A high-speed digital signal processor for atmospheric radar, part 7.3A

The Model SP-320 device is a monolithic realization of a complex general purpose signal processor, incorporating such features as a 32-bit ALU, a 16-bit x 16-bit combinatorial multiplier, and a 16-bit barrel shifter. The SP-320 is designed to operate as a slave processor to a host general purpose computer in applications such as coherent integration of a radar return signal in multiple ranges, or dedicated FFT processing. Presently available is an I/O module conforming to the Intel Multichannel interface standard; other I/O modules will be designed to meet specific user requirements. The main processor board includes input and output FIFO (First In First Out) memories, both with depths of 4096 W, to permit asynchronous operation between the source of data and the host computer. This design permits burst data rates in excess of 5 MW/s.

Brosnahan, J. W.↗

Robofurnace: A semi-automated laboratory chemical vapor deposition system for high-throughput nanomaterial synthesis and process discovery

Laboratory research and development on new materials, such as nanostructured thin films, often utilizes manual equipment such as tube furnaces due to its relatively low cost and ease of setup. However, these systems can be prone to inconsistent outcomes due to variations in standard operating procedures and limitations in performance such as heating and cooling rates restrict the parameter space that can be explored. Perhaps more importantly, maximization of research throughput and the successful and efficient translation of materials processing knowledge to production-scale systems, relies on the attainment of consistent outcomes. In response to this need, we present a semi-automated lab-scale chemical vapor deposition (CVD) furnace system, called “Robofurnace.” Robofurnace is an automated CVD system built around a standard tube furnace, which automates sample insertion and removal and uses motion of the furnace to achieve rapid heating and cooling. The system has a 10-sample magazine and motorized transfer arm, which isolates the samples from the lab atmosphere and enables highly repeatable placement of the sample within the tube. The system is designed to enable continuous operation of the CVD reactor, with asynchronous loading/unloading of samples. To demonstrate its performance, Robofurnace is used to develop a rapid CVD recipe for carbon nanotube (CNT) forest growth, achieving a 10-fold improvement in CNT forest mass density compared to a benchmark recipe using a manual tube furnace. In the long run, multiple systems like Robofurnace may be linked to share data among laboratories by methods such as Twitter. Our hope is Robofurnace and like automation will enable machine learning to optimize and discover relationships in complex material synthesis processes.

36 MATERIALS SCIENCE↗