Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Multiple processors”

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 667 records · Page 37

Comparing the OpenMP, MPI, and Hybrid Programming Paradigm on an SMP Cluster

Clusters of SMP (Symmetric Multi-Processors) nodes provide support for a wide range of parallel programming paradigms. The shared address space within each node is suitable for OpenMP parallelization. Message passing can be employed within and across the nodes of a cluster. Multiple levels of parallelism can be achieved by combining message passing and OpenMP parallelization. Which programming paradigm is the best will depend on the nature of the given problem, the hardware components of the cluster, the network, and the available software. In this study we compare the performance of different implementations of the same CFD benchmark application, using the same numerical algorithm but employing different programming paradigms.

Jost, Gabriele↗

Analyzing the Impact of System Reliability Events on Applications in the Titan Supercomputer

Extreme-scale computing systems employ Reliability, Availability and Serviceability (RAS) mechanisms and infrastructure to log events from multiple system components. In this paper, we analyze RAS logs in conjunction with the application placement and scheduling database, in order to understand the impact of common RAS events on application performance. This study conducted on the records of about 2 million applications executed on Titan supercomputer provides important insights for system users, operators and computer science researchers. Specifically, we investigate the impact of RAS events on application performance and its variability by comparing cases where events are recorded with corresponding cases where no events are recorded. Such a statistical investigation is possible since we observed that system users tend to execute their applications multiple times. Our analysis reveals that most RAS events do impact application performance, although not always. We also find that different system components affect application performance differently. In particular, our investigation includes the following components: parallel file system processor, memory, graphics processing units, system and user software issues. Our work establishes the importance of providing feedback to system users for increasing operational efficiency of extreme-scale systems.

Ashraf, Rizwan↗

[Activities of Bay Area Research Corporation]

During the final year of this effort the HALFSHEL code was converted to work on a fast single processor workstation from it s parallel configuration. This was done because NASA Ames NAS facility stopped supporting space science and we no longer had access to parallel computer time. The single processor version of HALFSHEL was upgraded to address low density cells by using a a 3-D SOR solver to solve the equation Delta central dot E = 0. We then upgraded the ionospheric load packages to provide a multiple species load of the ionosphere out to 1.4 Rm. With these new tools we began to perform a series of simulations to address the major topic of this research effort; determining the loss rate of O(sup +) and O2(sup +) from Mars. The simulations used the nominal Parker spiral field and in one case used a field perpendicular to the solar wind flow. The simulations were performed for three different solar EUV fluxes consistent with the different solar evolutionary states believed to exist before today. The 1 EUV case is the nominal flux of today. The 3 EUV flux is called Epoch 2 and has three times the flux of todays. The 6 EUV case is Epoch 3 and has 6 times the EUV flux of today.

Source record↗

Sample Processor for Life on Icy Worlds (SPLIce): Monolithic Manifold-Based System to Recover, Prepare, and Deliver Samples and Standards to Instrumentation Suites for Ocean World Life-Search Missions

A claim of life detection on one of the solar system’s icy ocean worlds would necessitate extraordinarily convincing evidence. Limited energy availability in the oceans of such bodies as Europa and Enceladus argues for microbes as most probable among possible life forms, but evidence of their existence in the surface layers of an icy moon or in a frozen plume ejected into space could take various forms, pointing to instrumentation suites as a preferred means to detect diverse molecular and morphological life indicators. Multiple disparate categories of positive detections could provide truly convincing evidence from samples that may be only a few micro-liters. SPLIce’s Foundation. Teams led by NASA’s Ames Research Center have developed and operated numerous small, live-biology and astrobiology science payloads in space over two decades. Since 2016, we have adapted and augmented their biological sample-handling systems to create compact, robust search-for-life fluidic processors designed to function after a decade or more in transit, in environments with very little gravity and lots of radiation: up to 100’s of kilorads.

Instrumentation suite↗

Flight Computer Processing Avionics for Space Station Microgravity Experiments: A Risk Assessment of Commercial Off-the-Shelf Utilization

NASA/Marshall Space Flight Center (MSFC) is continually looking for ways to reduce the costs and schedule and minimize the technical risks during the development of microgravity programs. One of the more prominent ways to minimize the cost and schedule is to use off-the-shelf hardware (OTS). However, the use of OTS often increases the risk. This paper addresses relevant factors considered during the selection and utilization of commercial off-the-shelf (COTS) flight computer processing equipment for the control of space station microgravity experiments. The paper will also discuss how to minimize the technical risks when using COTS processing hardware. Two microgravity experiments for which the COTS processing equipment is being evaluated for are the Equiaxed Dendritic Solidification Experiment (EDSE) and the Self-diffusion in Liquid Elements (SDLE) experiment. Since MSFC is the lead center for Microgravity research, EDSE and SDLE processor selection will be closely watched by other experiments that are being designed to meet payload carrier requirements. This includes the payload carriers planned for the International Space Station (ISS). The purpose of EDSE is to continue to investigate microstructural evolution of, and thermal interactions between multiple dendrites growing under diffusion controlled conditions. The purpose of SDLE is to determine accurate self-diffusivity data as a function of temperature for liquid elements selected as representative of class-like structures. In 1999 MSFC initiated a Center Director's Discretionary Fund (CDDF) effort to investigate and determine the optimal commercial data bus architecture that could lead to faster, better, and lower cost data acquisition systems for the control of microgravity experiments. As part of this effort various commercial data acquisition systems were acquired and evaluated. This included equipment with various form factors, (3U, 6U, others) and equipment that utilized various bus structures, (VME, PC104, STD bus). This evaluation of hardware was performed in conjunction with a trade study that considered over twenty (20) different factors relevant to the selection of an optimum design approach. These factors included; safety, sizing and timing, radiation hardness and single event upset, power consumption, heat dissipation, size and volume, expected service life, maintainability, heritage, operating systems, requirements for software reuse, availability of compatible interface boards, relative cost, schedule, reliability, EMI/EMC factors, "hot swap" capability, standards for conduction cooling, I/O capabilities, unique carrier requirements and operating system considerations. The approach to evaluate Safety as part of this study included a review of the Preliminary Hazard Analysis (PHA) for each of the experiment designs and a determination of how each hazard could be addressed and eliminated when different processors were selected. This included evaluating various design approaches and trade-offs between fault tolerant designs and fail-safe designs in accordance with NSTS 1700.7B. This will include the results of radiation testing where available. Various operating systems, such as VxWorks, Linux, QNX, and Embedded NT are evaluated and the advantages and disadvantages of their utilization are also addressed. Design implementation strategies for the various operating systems are considered and discussed. This paper presents the results and recommendations from this trade study. Preliminary conclusions from this study are that safety concerns from lack or radiation testing on COTS equipment can be addressed by additional testing and design considerations, the PC104 bus provided adequate I/O for the SDLE and EDSE microgravity experiments, and PC104 bus components offered significant advantages over VME and cPCI for weight and space reductions.

Estes, Howard↗

Parallel adaptation of general three-dimensional hybrid meshes

A new parallel dynamic mesh adaptation and load balancing algorithm for general hybrid grids has been developed. The meshes considered in this work are composed of four kinds of elements; tetrahedra, prisms, hexahedra and pyramids, which poses a challenge to parallel mesh adaptation. Additional complexity imposed by the presence of multiple types of elements affects especially data migration, updates of local data structures and interpartition data structures. Efficient partition of hybrid meshes has been accomplished by transforming them to suitable graphs and using serial graph partitioning algorithms. Communication among processors is based on the faces of the interpartition boundary and the termination detection algorithm of Dijkstra is employed to ensure proper flagging of edges for refinement. An inexpensive dynamic load balancing strategy is introduced to redistribute work load among processors after adaptation. In particular, only the initial coarse mesh, with proper weighting, is balanced which yields savings in computation time and relatively simple implementation of mesh quality preservation rules, while facilitating coarsening of refined elements. Special algorithms are employed for (i) data migration and dynamic updates of the local data structures, (ii) determination of the resulting interpartition boundary and (iii) identification of the communication pattern of processors. Several representative applications are included to evaluate the method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Laser-plasma source parameters for Kr, Gd, and Tb ions at 6.6 nm

There is increasing interest in extreme-ultraviolet (EUV) laser-based lamps for sub-10-nm lithography operating in the region of 6.6 nm. A collisional-radiative model is developed as a post-processor of a hydrodynamic code to investigate emission from resonance lines in Kr, Gd, and Tb ions under conditions typical for mass-limited EUV sources. The analysis reveals that maximum conversion efficiencies of Kr occur at 5 Multiplication-Sign 10{sup 10}W/cm{sup 2}, while for Gd and Tb it was Asymptotically-Equal-To 0.9%/2{pi}sr for laser intensities of (2-5) Multiplication-Sign 10{sup 12}W/cm{sup 2}.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Near-ground-state transport of trapped-ion qubits through a multidimensional array

We have demonstrated transport of {sup 9}Be{sup +}ions through a two-dimensional Paul-trap array that incorporates an X junction, while maintaining the ions near the motional ground state of the confining potential well. We expand on the first report of the experiment in Blakestad et al.[Phys. Rev. Lett. 102, 153002 (2009)], including a detailed discussion of how the transport potentials were calculated. Two main mechanisms that caused motional excitation during transport are explained, along with the methods used to mitigate such excitation. We reduced the motional excitation below the results in the above reference by a factor of approximately 50. The effect of a mu-metal shield on qubit coherence is also reported. Finally, we examined a method for exchanging energy between multiple motional modes on the few-quanta level, which could be useful for cooling motional modes without directly accessing the modes with lasers. These results establish how trapped ions can be transported in a large-scale quantum processor with high fidelity.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Dynamic Data-Driven SAR Image Reconstruction Using Multiple GPUs

According to the article, the reconstruction of nxn-pixel synthetic aperture radar (SAR) imagery using a backprojection algorithm incurs O(n 2 · m) cost, where m is the number of pulses. This paper presents dynamic data-driven multiresolution algorithms to speed up SAR backprojection on multiple graphics processing units (GPUs). A critical part of this spatially variant reconstruction process is load balancing, which circumvents asymmetric work assignment. Fine-tuned algorithms for GPUs are presented as a part of improving running time. Communication between processors is overlapped with GPU calculation to reduce communication time.

Engineering↗

DYCAST: A finite element program for the crash analysis of structures

DYCAST is a nonlinear structural dynamic finite element computer code developed for crash simulation. The element library contains stringers, beams, membrane skin triangles, plate bending triangles and spring elements. Changing stiffnesses in the structure are accounted for by plasticity and very large deflections. Material nonlinearities are accommodated by one of three options: elastic-perfectly plastic, elastic-linear hardening plastic, or elastic-nonlinear hardening plastic of the Ramberg-Osgood type. Geometric nonlinearities are handled in an updated Lagrangian formulation by reforming the structure into its deformed shape after small time increments while accumulating deformations, strains, and forces. The nonlinearities due to combined loadings are maintained, and stiffness variation due to structural failures are computed. Numerical time integrators available are fixed-step central difference, modified Adams, Newmark-beta, and Wilson-theta. The last three have a variable time step capability, which is controlled internally by a solution convergence error measure. Other features include: multiple time-load history tables to subject the structure to time dependent loading; gravity loading; initial pitch, roll, yaw, and translation of the structural model with respect to the global system; a bandwidth optimizer as a pre-processor; and deformed plots and graphics as post-processors.

Pifko, A. B.↗

The application of multiple instruction retry to VLIW architectures using compiler generated hazard-free code

In this paper, we describe the development of compiler assisted multiple instruction word retry for VLIW architectures. Compiler generated hazard-free code with different degrees of rollback capability is compacted by a trace scheduling algorithm. Performances are compared under three parameters: N, the rollback distance for uni-processors; P, the number of functional units; and Np, the rollback distance for VLIW architectures. In the majority of benchmarks examined, for fixed P and Np, the larger N tends to generate compacted code with better performance.

Chen, Shyh-Kwei↗

An evaluation of the CAN bus for use on the Europa Lander motor controller

This paper will present the application of the Controller Area Network (CAN) for communication between the Motor Control Processor and the Europa Lander Motor Control Cards. The Europa Lander preproject team is concerned about lowering the estimated mass, volume, and power of the mission concept so that it could maximize its science return. The motor controller uses multiple Motor Control Cards to control the 24 motors that will be operating in the extremes of the Europa environment. The CAN bus was considered because of its multi-drop nature, low power interface, and simple communication protocol.

Lias, Malcolm↗

FlexiWay: A Cache Energy Saving Technique Using Fine-grained Cache Reconfiguration

Recent trends of CMOS scaling and use of large last level caches (LLCs) have led to significant increase in the leakage energy consumption of LLCs and hence, managing their energy consumption has become extremely important in modern processor design. The conventional cache energy saving techniques require offline profiling or provide only coarse granularity of cache allocation. We present FlexiWay, a cache energy saving technique which uses dynamic cache reconfiguration. FlexiWay logically divides the cache sets into multiple (e.g. 16) modules and dynamically turns off suitable and possibly different number of cache ways in each module. FlexiWay has very small implementation overhead and it provides fine-grain cache allocation even with caches of typical associativity, e.g. an 8-way cache. Microarchitectural simulations have been performed using an x86-64 simulator and workloads from SPEC2006 suite. Also, FlexiWay has been compared with two conventional energy saving techniques. The results show that FlexiWay provides largest energy saving and incurs only small loss in performance. For single, dual and quad core systems, the average energy saving using FlexiWay are 26.2%, 25.7% and 22.4%, respectively.

Cache leakage energy saving↗

The data processor from the small scientific satellite

A reprogrammable data system aboard a small scientific satellite is described that samples and processes magnetospheric measurements for transmission to the ground. The lightweight configuration of the data system is made up of the program memory, data storage, input/output module, and a central processing unit. The system is designed for multiple missions.

Mccain, H. G.↗

Vector-Matrix Multiplication Engine for Neuromorphic Computation with a CBRAM Crossbar Array [Slides]

The core function of many neural network algorithms is the dot product, or vector matrix multiply (VMM) operation. Crossbar arrays utilizing resistive memory elements can reduce computational energy in neural algorithms by up to five orders of magnitude compared to conventional CPUs. Moving data between a processor, SRAM, and DRAM dominates energy consumption. By utilizing analog operations to reduce data movement, resistive memory crossbars can enable processing of large amounts of data at lower energy than conventional memory architectures.

97 MATHEMATICS AND COMPUTING↗

Real-space solution to the electronic structure problem for nearly a million electrons

We report a Kohn–Sham density functional theory calculation of a system with more than 200 000 atoms and 800 000 electrons using a real-space high-order finite-difference method to investigate the electronic structure of large spherical silicon nanoclusters. Our system of choice was a 20 nm large spherical nanocluster with 202 617 silicon atoms and 13 836 hydrogen atoms used to passivate the dangling surface bonds. To speed up the convergence of the eigenspace, we utilized Chebyshev-filtered subspace iteration, and for sparse matrix–vector multiplications, we used blockwise Hilbert space-filling curves, implemented in the PARSEC code. For this calculation, we also replaced our orthonormalization + Rayleigh–Ritz step with a generalized eigenvalue problem step. We utilized all of the 8192 nodes (458 752 processors) on the Frontera machine at the Texas Advanced Computing Center. We achieved two Chebyshev-filtered subspace iterations, yielding a good approximation of the electronic density of states. Our work pushes the limits on the capabilities of the current electronic structure solvers to nearly 106 electrons and demonstrates the potential of the real-space approach to efficiently parallelize large calculations on modern high-performance computing platforms.

Chemistry↗

Development of ROACH firmware for microwave multiplexed X-ray TES microcalorimeters

We are developing room temperature electronics based upon the ROACH platform for reading out microwave multiplexed X-ray TES. ROACH is an open-source hardware and software platform featuring a large Xilinx Field Programmable Gate Array (FPGA), Power PC processor, several 10GB Ethernet SFP+ interfaces, and a collection of daughter boards for analog signal generation and acquisition. The combination of a ROACH board, ADC/DAC conversion daughter boards, and hardware for RF mixing allows for the generation and capture of multiple RF tones for reading out microwave multiplexed x-ray TES microcalorimeters. The FPGA is used to generate multiple tones in base band, from 10MHz to 250MHz, which are subsequently mixed to RF in the multiple GHz range and sent through the microwave multiplexer. The tones are generated in the FPGA by storing a large lookup table in Quad Data Rate (QDR) SRAM modules and playing out the waveform to a DAC board. Once the signal has been modulated to RF, passed through the microwave multiplexer, and has been modulated back to base band, the signal is digitized by an ADC board. The tones are modulated to 0Hz by using a FPGA circuit consisting of a polyphase filter bank, several Xilinx FFT blocks, Xilinx CORDIC blocks (for converting to magnitude and phase), and special phase accumulator circuit for mixing to exactly 0Hz. Upwards of 256 channels can be simultaneously captured and written into a bank of 256 First-In-First-Out (FIFO) memories, with each FIFO corresponding to a channel. Individual channel data can be further processed in the FPGA before being streamed through a 10GB Ethernet fiber-optic interface to a Linux system. The Linux system runs software written in Python and QT C++ for controlling the ROACH system, capturing data, and processing data.

Firmware↗

Smart Power Supply for Battery-Powered Systems

A power supply for battery-powered systems has been designed with an embedded controller that is capable of monitoring and maintaining batteries, charging hardware, while maintaining output power. The power supply is primarily designed for rovers and other remote science and engineering vehicles, but it can be used in any battery alone, or battery and charging source applications. The supply can function autonomously, or can be connected to a host processor through a serial communications link. It can be programmed a priori or on the fly to return current and voltage readings to a host. It has two output power busses: a constant 24-V direct current nominal bus, and a programmable bus for output from approximately 24 up to approximately 50 V. The programmable bus voltage level, and its output power limit, can be changed on the fly as well. The power supply also offers options to reduce the programmable bus to 24 V when the set power limit is reached, limiting output power in the case of a system fault detected in the system. The smart power supply is based on an embedded 8051-type single-chip microcontroller. This choice was made in that a credible progression to flight (radiation hard, high reliability) can be assumed as many 8051 processors or gate arrays capable of accepting 8051-type core presently exist and will continue to do so for some time. To solve the problem of centralized control, this innovation moves an embedded microcontroller to the power supply and assigns it the task of overseeing the operation and charging of the power supply assets. This embedded processor is connected to the application central processor via a serial data link such that the central processor can request updates of various parameters within the supply, such as battery current, bus voltage, remaining power in battery estimations, etc. This supply has a direct connection to the battery bus for common (quiescent) power application. Because components from multiple vendors may have differing power needs, this supply also has a secondary power bus, which can be programmed a priori or on-the-fly to boost the primary battery voltage level from 24 to 50 V to accommodate various loads as they are brought on line. Through voltage and current monitoring, the device can also shield the charging source from overloads, keep it within safe operating modes, and can meter available power to the application and maintain safe operations.

Krasowski, Michael J.↗