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 649 records · Page 36

Microcomputer-based programmable optical signal processor

A microcomputer-based real-time programmable optical signal processing system utilizing a magneto-optic spatial light modulator (MOSLM) and a liquid crystal light valve (LCLV) is described. This system can perform a myriad of complicated optical operations, such as image correlation, image subtraction, and matrix multiplication. Its important asserts are its programmability and the capability of real-time addressing. These are important for telerobotic vision in space automation applications. Design specifications and suggestions for practical implementation of the system are discussed. Some preliminary experimental demonstrations are conducted to demonstrate applications of the proposed system to image correlation for optical pattern recognition, image subtraction for lC chip inspection, and matrix multiplication for optical computing.

Yu, F. T. S.↗

pGraph: Efficient Parallel Construction of Large-Scale Protein Sequence Homology Graphs

Detecting sequence homology between protein sequences is a fundamental problem in computational molecular biology, with a pervasive application in nearly all analyses that aim to structurally and functionally characterize protein molecules. While detecting the homology between two protein sequences is relatively inexpensive, detecting pairwise homology for a large number of protein sequences can become computationally prohibitive for modern inputs, often requiring millions of CPU hours. Yet, there is currently no robust support to parallelize this kernel. In this paper, we identify the key characteristics that make this problemparticularly hard to parallelize, and then propose a new parallel algorithm that is suited for detecting homology on large data sets using distributed memory parallel computers. Our method, called pGraph, is a novel hybrid between the hierarchical multiple-master/worker model and producer-consumer model, and is designed to break the irregularities imposed by alignment computation and work generation. Experimental results show that pGraph achieves linear scaling on a 2,048 processor distributed memory cluster for a wide range of inputs ranging from as small as 20,000 sequences to 2,560,000 sequences. In addition to demonstrating strong scaling, we present an extensive report on the performance of the various system components and related parametric studies.

Environmental Molecular Sciences Laboratory↗

Parallel Application Performance on Two Generations of Intel Xeon HPC Platforms

Two next-generation node configurations hosting the Haswell microarchitecture were tested with a suite of microbenchmarks and application examples, and compared with a current Ivy Bridge production node on NREL" tm s Peregrine high-performance computing cluster. A primary conclusion from this study is that the additional cores are of little value to individual task performance--limitations to application parallelism, or resource contention among concurrently running but independent tasks, limits effective utilization of these added cores. Hyperthreading generally impacts throughput negatively, but can improve performance in the absence of detailed attention to runtime workflow configuration. The observations offer some guidance to procurement of future HPC systems at NREL. First, raw core count must be balanced with available resources, particularly memory bandwidth. Balance-of-system will determine value more than processor capability alone. Second, hyperthreading continues to be largely irrelevant to the workloads that are commonly seen, and were tested here, at NREL. Finally, perhaps the most impactful enhancement to productivity might occur through enabling multiple concurrent jobs per node. Given the right type and size of workload, more may be achieved by doing many slow things at once, than fast things in order.

97 MATHEMATICS AND COMPUTING↗

Solving large sparse eigenvalue problems on supercomputers

An important problem in scientific computing consists in finding a few eigenvalues and corresponding eigenvectors of a very large and sparse matrix. The most popular methods to solve these problems are based on projection techniques on appropriate subspaces. The main attraction of these methods is that they only require the use of the matrix in the form of matrix by vector multiplications. The implementations on supercomputers of two such methods for symmetric matrices, namely Lanczos' method and Davidson's method are compared. Since one of the most important operations in these two methods is the multiplication of vectors by the sparse matrix, methods of performing this operation efficiently are discussed. The advantages and the disadvantages of each method are compared and implementation aspects are discussed. Numerical experiments on a one processor CRAY 2 and CRAY X-MP are reported. Possible parallel implementations are also discussed.

Philippe, Bernard↗

Universal linear intensity transformations using spatially incoherent diffractive processors

Abstract Under spatially coherent light, a diffractive optical network composed of structured surfaces can be designed to perform any arbitrary complex-valued linear transformation between its input and output fields-of-view (FOVs) if the total number ( N ) of optimizable phase-only diffractive features is ≥~2 N i N o , where N i and N o refer to the number of useful pixels at the input and the output FOVs, respectively. Here we report the design of a spatially incoherent diffractive optical processor that can approximate any arbitrary linear transformation in time-averaged intensity between its input and output FOVs. Under spatially incoherent monochromatic light, the spatially varying intensity point spread function ( H ) of a diffractive network, corresponding to a given, arbitrarily-selected linear intensity transformation, can be written as H ( m , n ; m ′, n ′) = | h ( m , n ; m ′, n ′)| 2 , where h is the spatially coherent point spread function of the same diffractive network, and ( m , n ) and ( m ′, n ′) define the coordinates of the output and input FOVs, respectively. Using numerical simulations and deep learning, supervised through examples of input-output profiles, we demonstrate that a spatially incoherent diffractive network can be trained to all-optically perform any arbitrary linear intensity transformation between its input and output if N ≥ ~2 N i N o . We also report the design of spatially incoherent diffractive networks for linear processing of intensity information at multiple illumination wavelengths, operating simultaneously. Finally, we numerically demonstrate a diffractive network design that performs all-optical classification of handwritten digits under spatially incoherent illumination, achieving a test accuracy of >95%. Spatially incoherent diffractive networks will be broadly useful for designing all-optical visual processors that can work under natural light.

36 MATERIALS SCIENCE↗

A Flexible-blocking Based Approach for Performance Tuning of Matrix Multiplication Routines for Large Matrices with Edge Cases

Efficient and scalable matrix operations are being highly demanding in the recent era of Machine Learning, Deep Learning, and Big Data Analytics. The two commonly used matrix-matrix operations in the Basic Linear Algebra Subprograms (BLAS) specification are General Matrix-Matrix multiplication (GEMM) and Symmetric Rank-k update (SYRK). The SYRK routine is a specialization of the GEMM routine, where half of the multiplications are skipped as the resultant matrix is known to be symmetric. Fortunately, several linear algebra libraries implement these BLAS routines quite efficiently. The libraries usually partition the input matrices into blocks and place them in processor caches, thus improving performance by leveraging the caches. However, the contemporary libraries are highly optimized for squarish matrices, but the performance degrades significantly for the matrices with edge case (strictly thin or strictly fat shapes) in the multicore machine. The primary reason is that the current state-of-the-art libraries make fixed block shapes based on a processor architecture, and do not consider the shape of the input matrices. In this paper, we propose a new blocking approach, we name it Flexible-blocking, to mitigate the scalability issues. In contrast to the contemporary libraries, our approach formulates the blocks of the input matrices based on the shapes of the matrices as well as the number of threads used in the implementation. Our proposed technique shows noticeable performance improvement on multicore shared-memory machines for the edge case matrices.

Hossain, Md Mosharaf↗

GMH: A Message Passing Toolkit for GPU Clusters

Driven by the market demand for high-definition 3D graphics, commodity graphics processing units (GPUs) have evolved into highly parallel, multi-threaded, many-core processors, which are ideal for data parallel computing. Many applications have been ported to run on a single GPU with tremendous speedups using general C-style programming languages such as CUDA. However, large applications require multiple GPUs and demand explicit message passing. This paper presents a message passing toolkit, called GMH (GPU Message Handler), on NVIDIA GPUs. This toolkit utilizes a data-parallel thread group as a way to map multiple GPUs on a single host to an MPI rank, and introduces a notion of virtual GPUs as a way to bind a thread to a GPU automatically. This toolkit provides high performance MPI style point-to-point and collective communication, but more importantly, facilitates event-driven APIs to allow an application to be managed and executed by the toolkit at runtime.

99 GENERAL AND MISCELLANEOUS↗

Imaging radar polarimetry from wave synthesis

A new approach is reported to the measurement of the complete polarization signature of each resolution element in an image implemented with an airborne synthetic aperture radar system. Signals recorded on one data pass from orthogonal linearly polarized antennas are utilized. The signals are combined in a data processor to synthesize any desired combination of transmit and receive polarizations. The technique permits measurement of the complex, multichannel reflectivity of a scene on a single aircraft pass and to late reprocess the data to provide multiple image maps, with each representing the backscattered energy from the scene measured with a different combination of observational transmit and recieve polarizations. The resulting polarization signature measurements indicate optimum polarizations for observations of certain classes of objects and give insight into the identification of dominant scattering mechanisms for each kind of object. The mathematical model for polarization synthesis is summarized, and some theoretical polarization measurements are illustrated for several types of targets. The overall radar system implementation is described in detail. Some analyses of data acquired on three aircraft flights are presented. The technique has been applied to mapping and differentiation of lava flows and to differentiation of forested and clear-cut areas.

Zebker, Howard A.↗

Particle displacement tracking applied to air flows

Electronic Particle Image Velocimeter (PIV) techniques offer many advantages over conventional photographic PIV methods such as fast turn around times and simplified data reduction. A new all electronic PIV technique was developed which can measure high speed gas velocities. The Particle Displacement Tracking (PDT) technique employs a single cw laser, small seed particles (1 micron), and a single intensified, gated CCD array frame camera to provide a simple and fast method of obtaining two-dimensional velocity vector maps with unambiguous direction determination. Use of a single CCD camera eliminates registration difficulties encountered when multiple cameras are used to obtain velocity magnitude and direction information. An 80386 PC equipped with a large memory buffer frame-grabber board provides all of the data acquisition and data reduction operations. No array processors of other numerical processing hardware are required. Full video resolution (640x480 pixel) is maintained in the acquired images, providing high resolution video frames of the recorded particle images. The time between data acquisition to display of the velocity vector map is less than 40 sec. The new electronic PDT technique is demonstrated on an air nozzle flow with velocities less than 150 m/s.

Wernet, Mark P.↗

Rainbow net analysis of VAXcluster system availability

A system modeling technique, Rainbow Nets, is used to evaluate the availability and mean-time-to-interrupt of the VAXcluster. These results are compared to the exact analytic results showing that reasonable accuracy is achieved through simulation. The complexity of the Rainbow Net does not increase as the number of processors increases, but remains constant, unlike a Markov model which expands exponentially. The constancy is achieved by using tokens with identity attributes (items) that can have additional attributes associated with them (features) which can exist in multiple states. The time to perform the simulation increases, but this is a polynomial increase rather than exponential. There is no restriction on distributions used for transition firing times, allowing real situations to be modeled more accurately by choosing the distribution which best fits the system performance and eliminating the need for simplifying assumptions.

Johnson, Allen M., Jr.↗

A distributed agent architecture for real-time knowledge-based systems: Real-time expert systems project, phase 1

We propose a distributed agent architecture (DAA) that can support a variety of paradigms based on both traditional real-time computing and artificial intelligence. DAA consists of distributed agents that are classified into two categories: reactive and cognitive. Reactive agents can be implemented directly in Ada to meet hard real-time requirements and be deployed on on-board embedded processors. A traditional real-time computing methodology under consideration is the rate monotonic theory that can guarantee schedulability based on analytical methods. AI techniques under consideration for reactive agents are approximate or anytime reasoning that can be implemented using Bayesian belief networks as in Guardian. Cognitive agents are traditional expert systems that can be implemented in ART-Ada to meet soft real-time requirements. During the initial design of cognitive agents, it is critical to consider the migration path that would allow initial deployment on ground-based workstations with eventual deployment on on-board processors. ART-Ada technology enables this migration while Lisp-based technologies make it difficult if not impossible. In addition to reactive and cognitive agents, a meta-level agent would be needed to coordinate multiple agents and to provide meta-level control.

Lee, S. Daniel↗

Particle displacement tracking applied to air flows

Electronic Particle Image Velocimetric (PIV) techniques offer many advantages over conventional photographic PIV methods such as fast turn around times and simplified data reduction. A new all electronic PIV technique was developed which can measure high speed gas velocities. The Particle Displacement Tracking (PDT) technique employs a single CW laser, small seed particles (1 micron), and a single intensified, gated CCD array frame camera to provide a simple and fast method of obtaining two-dimensional velocity vector maps with unambiguous direction determination. Use of a single CCD camera eliminates registration difficulties encountered when multiple cameras are used to obtain velocity magnitude and direction information. An 80386 PC equipped with a large memory buffer frame-grabber board provides all of the data acquisition and data reduction operations. No array processors of other numerical processing hardware are required. Full video resolution (640 x 480 pixel) is maintained in the acquired images, providing high resolution video frames of the recorded particle images. The time between data acquisition to display of the velocity vector map is less than 40 sec. The new electronic PDT technique is demonstrated on an air nozzle flow with velocities less than 150 m/s.

Wernet, Mark P.↗

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↗

Big Science in Small Packages: a Constellation of Cubesats to Search for Near Earth Asteroids

Astronomers for many decades have been searching for Near Earth Asteroids (NEA) that could potentially impact the Earth in the future. Large Synoptic Survey Telescope (LSST), when completed, would be by far the most powerful tool for a ground-based search for NEAs. In space, IR telescopes are undergoing detailed design studies. These facilities are projected to find 90% of potentially hazardous NEAs in about a dozen years. We introduce a different approach to search for these objects by using multiple micro- and nano-spacecraft distributed in orbit around the Sun. Their ability to faint moving objects with small telescopes is based on a new generation of low noise, fast-frame cameras combined with processors that can execute 100’s of gigaflops in a low power space environment.

Shao, Michael↗

Software Defined GPS Receiver for International Space Station

JPL is providing a software defined radio (SDR) that will fly on the International Space Station (ISS) as part of the CoNNeCT project under NASA's SCaN program. The SDR consists of several modules including a Baseband Processor Module (BPM) and a GPS Module (GPSM). The BPM executes applications (waveforms) consisting of software components for the embedded SPARC processor and logic for two Virtex II Field Programmable Gate Arrays (FPGAs) that operate on data received from the GPSM. GPS waveforms on the SDR are enabled by an L-Band antenna, low noise amplifier (LNA), and the GPSM that performs quadrature downconversion at L1, L2, and L5. The GPS waveform for the JPL SDR will acquire and track L1 C/A, L2C, and L5 GPS signals from a CoNNeCT platform on ISS, providing the best GPS-based positioning of ISS achieved to date, the first use of multiple frequency GPS on ISS, and potentially the first L5 signal tracking from space. The system will also enable various radiometric investigations on ISS such as local multipath or ISS dynamic behavior characterization. In following the software-defined model, this work will create a highly portable GPS software and firmware package that can be adapted to another platform with the necessary processor and FPGA capability. This paper also describes ISS applications for the JPL CoNNeCT SDR GPS waveform, possibilities for future global navigation satellite system (GNSS) tracking development, and the applicability of the waveform components to other space navigation applications.

software defined radio (SDR)↗

CrossSim Inference Manual v2.0

Neural networks are largely based on matrix computations. During forward inference, the most heavily used compute kernel is the matrix-vector multiplication (MVM): $W \vec{x} $. Inference is a first frontier for the deployment of next-generation hardware for neural network applications, as it is more readily deployed in edge devices, such as mobile devices or embedded processors with size, weight, and power constraints. Inference is also easier to implement in analog systems than training, which has more stringent device requirements. The main processing kernel used during inference is the MVM.

97 MATHEMATICS AND COMPUTING↗

Development and evaluation of a fault-tolerant multiprocessor (FTMP) computer. Volume 1: FTMP principles of operation

The basic organization of the fault tolerant multiprocessor, (FTMP) is that of a general purpose homogeneous multiprocessor. Three processors operate on a shared system (memory and I/O) bus. Replication and tight synchronization of all elements and hardware voting is employed to detect and correct any single fault. Reconfiguration is then employed to repair a fault. Multiple faults may be tolerated as a sequence of single faults with repair between fault occurrences.

Smith, T. B., Jr.↗

Description and Simulation of a Fast Packet Switch Architecture for Communication Satellites

The NASA Lewis Research Center has been developing the architecture for a multichannel communications signal processing satellite (MCSPS) as part of a flexible, low-cost meshed-VSAT (very small aperture terminal) network. The MCSPS architecture is based on a multifrequency, time-division-multiple-access (MF-TDMA) uplink and a time-division multiplex (TDM) downlink. There are eight uplink MF-TDMA beams, and eight downlink TDM beams, with eight downlink dwells per beam. The information-switching processor, which decodes, stores, and transmits each packet of user data to the appropriate downlink dwell onboard the satellite, has been fully described by using VHSIC (Very High Speed Integrated-Circuit) Hardware Description Language (VHDL). This VHDL code, which was developed in-house to simulate the information switching processor, showed that the architecture is both feasible and viable. This paper describes a shared-memory-per-beam architecture, its VHDL implementation, and the simulation efforts.

Quintana, Jorge A.↗