Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “processor performance factors”

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 73 records · Page 4

Performance of a parallel algebraic multilevel preconditioner for stabilized finite element semiconductor device modeling

In this study results are presented for the large-scale parallel performance of an algebraic multilevel preconditioner for solution of the drift-diffusion model for semiconductor devices. The preconditioner is the key numerical procedure determining the robustness, efficiency and scalability of the fully-coupled Newton-Krylov based, nonlinear solution method that is employed for this system of equations. The coupled system is comprised of a source term dominated Poisson equation for the electric potential, and two convection-diffusion-reaction type equations for the electron and hole concentration. The governing PDEs are discretized in space by a stabilized finite element method. Solution of the discrete system is obtained through a fully-implicit time integrator, a fully-coupled Newton-based nonlinear solver, and a restarted GMRES Krylov linear system solver. The algebraic multilevel preconditioner is based on an aggressive coarsening graph partitioning of the nonzero block structure of the Jacobian matrix. Representative performance results are presented for various choices of multigrid V-cycles and W-cycles and parameter variations for smoothers based on incomplete factorizations. Parallel scalability results are presented for solution of up to 10{sup 8} unknowns on 4096 processors of a Cray XT3/4 and an IBM POWER eServer system.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Task Parallel Incomplete Cholesky Factorization using 2D Partitioned-Block Layout

We introduce a task-parallel algorithm for sparse incomplete Cholesky factorization that utilizes a 2D sparse partitioned-block layout of a matrix. Our factorization algorithm follows the idea of algorithms-by-blocks by using the block layout. The algorithm-byblocks approach induces a task graph for the factorization. These tasks are inter-related to each other through their data dependences in the factorization algorithm. To process the tasks on various manycore architectures in a portable manner, we also present a portable tasking API that incorporates different tasking backends and device-specific features using an open-source framework for manycore platforms i.e., Kokkos. A performance evaluation is presented on both Intel Sandybridge and Xeon Phi platforms for matrices from the University of Florida sparse matrix collection to illustrate merits of the proposed task-based factorization. Experimental results demonstrate that our task-parallel implementation delivers about 26.6x speedup (geometric mean) over single-threaded incomplete Choleskyby- blocks and 19.2x speedup over serial Cholesky performance which does not carry tasking overhead using 56 threads on the Intel Xeon Phi processor for sparse matrices arising from various application problems.

2D layout↗

Sensor Data Processing for Tracking Underwater Threats Using Terascale Optical Core Devices

A critical aspect of littoral surveillance (including port protection) involves the localization and tracking of underwater threats such as manned or unmanned autonomous underwater vehicles. In this article, we present a methodology for locating underwater threat sources from uncertain sensor network data, and illustrate the threat tracking aspects using active sonars in a matched filter framework. The novelty of the latter paradigm lies in its implementation on a tera-scale optical core processor, EnLight , recently introduced by Lenslet Laboratories. This processor is optimized for array operations, which it performs in a fixed point arithmetic architecture at tera-scale throughput. Using the EnLight 64 prototype processor, our results (i) illustrate the ability to reach a robust tracking accuracy, and (ii) demonstrate that a considerable speed-up (a factor of over 13,000) can be achieved when compared to an Intel XeonTM processor in the computation of sets of 80K-sample complex Fourier transforms that are associated with our matched filter techniques.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

A communication-avoiding, hybrid-parallel, rank-revealing orthogonalization method.

Orthogonalization consumes much of the run time of many iterative methods for solving sparse linear systems and eigenvalue problems. Commonly used algorithms, such as variants of Gram-Schmidt or Householder QR, have performance dominated by communication. Here, 'communication' includes both data movement between the CPU and memory, and messages between processors in parallel. Our Tall Skinny QR (TSQR) family of algorithms requires asymptotically fewer messages between processors and data movement between CPU and memory than typical orthogonalization methods, yet achieves the same accuracy as Householder QR factorization. Furthermore, in block orthogonalizations, TSQR is faster and more accurate than existing approaches for orthogonalizing the vectors within each block ('normalization'). TSQR's rank-revealing capability also makes it useful for detecting deflation in block iterative methods, for which existing approaches sacrifice performance, accuracy, or both. We have implemented a version of TSQR that exploits both distributed-memory and shared-memory parallelism, and supports real and complex arithmetic. Our implementation is optimized for the case of orthogonalizing a small number (5-20) of very long vectors. The shared-memory parallel component uses Intel's Threading Building Blocks, though its modular design supports other shared-memory programming models as well, including computation on the GPU. Our implementation achieves speedups of 2 times or more over competing orthogonalizations. It is available now in the development branch of the Trilinos software package, and will be included in the 10.8 release.

99 GENERAL AND MISCELLANEOUS↗

Development and modification of a digital program for final approach to landing

The development and implementation of a dynamic digital computer simulator which may be used to evaluate aircraft performance when operating under the control and guidance of various navigation, landing, and flight control systems are discussed. The digital computer program may be used to simulate and evaluate the relationships and interactions between various factors such as the microwave landing system, avionics receivers and onboard processors, aircraft aerodynamics, aircraft automatic control systems, control surfaces, and wind and other external effects. The models used to represent aircraft aerodynamics, control system and control surfaces; weather and wind effects; and the microwave landing system are described. Example results are presented for a simulation of a Boeing 737 using two sample control systems while subjected to various atmospheric conditions and microwave landing system errors. The limitations and performance capabilities of these control systems are discussed in terms of their ability to utilize the microwave landing system signal.

Duff, W. G.↗

An Efficient Scheme for Updating Sparse Cholesky Factors

Raghavan had earlier developed the software package DCSPACK which can be used for solving sparse linear systems where the coefficient matrix is symmetric and positive definite (this project was not funded by NASA but by agencies such as NSF). DSCPACK-S is the serial code and DSCPACK-P is a parallel implementation suitable for multiprocessors or networks-of-workstations with message passing using MCI. The main algorithm used is the Cholesky factorization of a sparse symmetric positive positive definite matrix A = LL(T). The code can also compute the factorization A = LDL(T). The complexity of the software arises from several factors relating to the sparsity of the matrix A. A sparse N x N matrix A has typically less that cN nonzeroes where c is a small constant. If the matrix were dense, it would have O(N2) nonzeroes. The most complicated part of such sparse Cholesky factorization relates to fill-in, i.e., zeroes in the original matrix that become nonzeroes in the factor L. An efficient implementation depends to a large extent on complex data structures and on techniques from graph theory to reduce, identify, and manage fill. DSCPACK is based on an efficient multifrontal implementation with fill-managing algorithms and implementation arising from earlier research by Raghavan and others. Sparse Cholesky factorization is typically a four step process: (1) ordering to compute a fill-reducing numbering, (2) symbolic factorization to determine the nonzero structure of L, (3) numeric factorization to compute L, and, (4) triangular solution to solve L(T)x = y and Ly = b. The first two steps are symbolic and are performed using the graph of the matrix. The numeric factorization step is of dominant cost and there are several schemes for improving performance by exploiting the nested and dense structure of groups of columns in the factor. The latter are aimed at better utilization of the cache-memory hierarchy on modem processors to prevent cache-misses and provide execution rates (operations/second) that are close to the peak rates for dense matrix computations. Currently, EPISCOPACY is being used in an application at NASA directed by J. Newman and M. James. We propose the implementation of efficient schemes for updating the LL(T) or LDL(T) factors computed in DSCPACK-S to meet the computational requirements of their project. A brief description is provided in the next section.

Raghavan, Padma↗

The cost of conservative synchronization in parallel discrete event simulations

The performance of a synchronous conservative parallel discrete-event simulation protocol is analyzed. The class of simulation models considered is oriented around a physical domain and possesses a limited ability to predict future behavior. A stochastic model is used to show that as the volume of simulation activity in the model increases relative to a fixed architecture, the complexity of the average per-event overhead due to synchronization, event list manipulation, lookahead calculations, and processor idle time approach the complexity of the average per-event overhead of a serial simulation. The method is therefore within a constant factor of optimal. The analysis demonstrates that on large problems--those for which parallel processing is ideally suited--there is often enough parallel workload so that processors are not usually idle. The viability of the method is also demonstrated empirically, showing how good performance is achieved on large problems using a thirty-two node Intel iPSC/2 distributed memory multiprocessor.

Nicol, David M.↗

Noiseless coding for the Gamma Ray spectrometer

The payload of several future unmanned space missions will include a sophisticated gamma ray spectrometer. Severely constrained data rates during certain portions of these missions could limit the possible science return from this instrument. This report investigates the application of universal noiseless coding techniques to represent gamma ray spectrometer data more efficiently without any loss in data integrity. Performance results demonstrate compression factors from 2.5:1 to 20:1 in comparison to a standard representation. Feasibility was also demonstrated by implementing a microprocessor breadboard coder/decoder using an Intel 8086 processor.

Rice, R.↗

A report documenting the completion of the Los Alamos National Laboratory portion of the ASC level II milestone ""Visualization on the supercomputing platform

This report provides documentation for the completion of the Los Alamos portion of the ASC Level II 'Visualization on the Supercomputing Platform' milestone. This ASC Level II milestone is a joint milestone between Sandia National Laboratory and Los Alamos National Laboratory. The milestone text is shown in Figure 1 with the Los Alamos portions highlighted in boldfaced text. Visualization and analysis of petascale data is limited by several factors which must be addressed as ACES delivers the Cielo platform. Two primary difficulties are: (1) Performance of interactive rendering, which is the most computationally intensive portion of the visualization process. For terascale platforms, commodity clusters with graphics processors (GPUs) have been used for interactive rendering. For petascale platforms, visualization and rendering may be able to run efficiently on the supercomputer platform itself. (2) I/O bandwidth, which limits how much information can be written to disk. If we simply analyze the sparse information that is saved to disk we miss the opportunity to analyze the rich information produced every timestep by the simulation. For the first issue, we are pursuing in-situ analysis, in which simulations are coupled directly with analysis libraries at runtime. This milestone will evaluate the visualization and rendering performance of current and next generation supercomputers in contrast to GPU-based visualization clusters, and evaluate the perfromance of common analysis libraries coupled with the simulation that analyze and write data to disk during a running simulation. This milestone will explore, evaluate and advance the maturity level of these technologies and their applicability to problems of interest to the ASC program. In conclusion, we improved CPU-based rendering performance by a a factor of 2-10 times on our tests. In addition, we evaluated CPU and CPU-based rendering performance. We encourage production visualization experts to consider using CPU-based rendering solutions when it is appropriate. For example, on remote supercomputers CPU-based rendering can offer a means of viewing data without having to offload the data or geometry onto a CPU-based visualization system. In terms of comparative performance of the CPU and CPU we believe that further optimizations of the performance of both CPU or CPU-based rendering are possible. The simulation community is currently confronting this reality as they work to port their simulations to different hardware architectures. What is interesting about CPU rendering of massive datasets is that for part two decades CPU performance has significantly outperformed CPU-based systems. Based on our advancements, evaluations and explorations we believe that CPU-based rendering has returned as one viable option for the visualization of massive datasets.

97 MATHEMATICS AND COMPUTING↗

Single-Chip FPGA Azimuth Pre-Filter for SAR

A field-programmable gate array (FPGA) on a single lightweight, low-power integrated-circuit chip has been developed to implement an azimuth pre-filter (AzPF) for a synthetic-aperture radar (SAR) system. The AzPF is needed to enable more efficient use of data-transmission and data-processing resources: In broad terms, the AzPF reduces the volume of SAR data by effectively reducing the azimuth resolution, without loss of range resolution, during times when end users are willing to accept lower azimuth resolution as the price of rapid access to SAR imagery. The data-reduction factor is selectable at a decimation factor, M, of 2, 4, 8, 16, or 32 so that users can trade resolution against processing and transmission delays. In principle, azimuth filtering could be performed in the frequency domain by use of fast-Fourier-transform processors. However, in the AzPF, azimuth filtering is performed in the time domain by use of finite-impulse-response filters. The reason for choosing the time-domain approach over the frequency-domain approach is that the time-domain approach demands less memory and a lower memory-access rate. The AzPF operates on the raw digitized SAR data. The AzPF includes a digital in-phase/quadrature (I/Q) demodulator. In general, an I/Q demodulator effects a complex down-conversion of its input signal followed by low-pass filtering, which eliminates undesired sidebands. In the AzPF case, the I/Q demodulator takes offset video range echo data to the complex baseband domain, ensuring preservation of signal phase through the azimuth pre-filtering process. In general, in an SAR I/Q demodulator, the intermediate frequency (fI) is chosen to be a quarter of the range-sampling frequency and the pulse-repetition frequency (fPR) is chosen to be a multiple of fI. The AzPF also includes a polyphase spatial-domain pre-filter comprising four weighted integrate-and-dump filters with programmable decimation factors and overlapping phases. To prevent aliasing of signals, the bandwidth of the AzPF is made 80 percent of fPR/M. The choice of four as the number of overlapping phases is justified by prior research in which it was shown that a filter of length 4M can effect an acceptable transfer function. The figure depicts prototype hardware comprising the AzPF and ancillary electronic circuits. The hardware was found to satisfy performance requirements in real-time tests at a sampling rate of 100 MHz.

Gudim, Mimi↗

A scalable geometric multigrid solver for nonsymmetric elliptic systems with application to variable-density flows

A geometric multigrid algorithm is introduced for solving nonsymmetric linear systems resulting from the discretization of the variable density Navier–Stokes equations on nonuniform structured rectilinear grids and high-Reynolds number flows. The restriction operation is defined such that the resulting system on the coarser grids is symmetric, thereby allowing for the use of efficient smoother algorithms. To achieve an optimal rate of convergence, the sequence of interpolation and restriction operations are determined through a dynamic procedure. A parallel partitioning strategy is introduced to minimize communication while maintaining the load balance between all processors. To test the proposed algorithm, we consider two cases: 1) homogeneous isotropic turbulence discretized on uniform grids and 2) turbulent duct flow discretized on stretched grids. Testing the algorithm on systems with up to a billion unknowns shows that the cost varies linearly with the number of unknowns. This O (N) behavior confirms the robustness of the proposed multigrid method regarding ill-conditioning of large systems characteristic of multiscale high-Reynolds number turbulent flows. The robustness of our method to density variations is established by considering cases where density varies sharply in space by a factor of up to 104, showing its applicability to two-phase flow problems. Strong and weak scalability studies are carried out, employing up to 30,000 processors, to examine the parallel performance of our implementation. Excellent scalability of our solver is shown for a granularity as low as 104 to 105 unknowns per processor. At its tested peak throughput, it solves approximately 4 billion unknowns per second employing over 16,000 processors with a parallel efficiency higher than 50%.

Computer Science↗

Mapping between parallel processor structures and programs

This paper reports some ongoing research efforts at Stanford in allocation of parallel processing resources. Both processor structures and program structures have their own characteristics. Resource allocation binds the two structures during program execution. The mapping problem determines what processor structure and program structure may be combined to obtain maximum speedup. Three approaches to this mapping problem are considered. Two important factors, granularity and interaction delay, are also considered. A new hierarchical approach to structure definition is outlined. Effective and efficient tools are necessary for the study of the mapping problem. A fast turn-around simulation environment developed for investigating partition strategies for distributed computations and a computationally efficient method to predict performance of parallel processor structures are described.

Ngai, Tin-Fook↗

The Performance of the NAS HSPs in 1st Half of 1994

During the first six months of 1994, the NAS (National Airspace System) 16-CPU Y-MP C90 Von Neumann (VN) delivered an average throughput of 4.045 GFLOPS while the ACSF (Aeronautics Consolidated Supercomputer Facility) 8-CPU Y-MP C90 Eagle averaged 1.658 GFLOPS. The VN rate represents a machine efficiency of 26.3% whereas the Eagle rate corresponds to a machine efficiency of 21.6%. VN displayed a greater efficiency than Eagle primarily because the stronger workload demand for its CPU cycles allowed it to devote more time to user programs and less time to idle. An additional factor increasing VN efficiency was the ability of the UNICOS 8.0 Operating System to deliver a larger fraction of CPU time to user programs. Although measurements indicate increasing vector length for both workloads, insufficient vector lengths continue to hinder HSP (High Speed Processor) performance. To improve HSP performance, NAS should continue to encourage the HSP users to modify their codes to increase program vector length.

Bergeron, Robert J.↗

Noiseless coding for the magnetometer

Future unmanned space missions will continue to seek a full understanding of magnetic fields throughout the solar system. Severely constrained data rates during certain portions of these missions could limit the possible science return. This publication investigates the application of universal noiseless coding techniques to more efficiently represent magnetometer data without any loss in data integrity. Performance results indicated that compression factors of 2:1 to 6:1 can be expected. Feasibility for general deep space application was demonstrated by implementing a microprocessor breadboard coder/decoder using the Intel 8086 processor. The Comet Rendezvous Asteroid Flyby mission will incorporate these techniques in a buffer feedback, rate-controlled configuration. The characteristics of this system are discussed.

Rice, Robert F.↗

SpaceCube Version 1.5

SpaceCube 1.5 is a high-performance and low-power system in a compact form factor. It is a hybrid processing system consisting of CPU (central processing unit), FPGA (field-programmable gate array), and DSP (digital signal processor) processing elements. The primary processing engine is the Virtex- 5 FX100T FPGA, which has two embedded processors. The SpaceCube 1.5 System was a bridge to the SpaceCube 2.0 and SpaceCube 2.0 Mini processing systems. The SpaceCube 1.5 system was the primary avionics in the successful SMART (Small Rocket/Spacecraft Technology) Sounding Rocket mission that was launched in the summer of 2011. For SMART and similar missions, an avionics processor is required that is reconfigurable, has high processing capability, has multi-gigabit interfaces, is low power, and comes in a rugged/compact form factor. The original SpaceCube 1.0 met a number of the criteria, but did not possess the multi-gigabit interfaces that were required and is a higher-cost system. The SpaceCube 1.5 was designed with those mission requirements in mind. The SpaceCube 1.5 features one Xilinx Virtex-5 FX100T FPGA and has excellent size, weight, and power characteristics [4×4×3 in. (approx. = 10×10×8 cm), 3 lb (approx. = 1.4 kg), and 5 to 15 W depending on the application]. The estimated computing power of the two PowerPC 440s in the Virtex-5 FPGA is 1100 DMIPS each. The SpaceCube 1.5 includes two Gigabit Ethernet (1 Gbps) interfaces as well as two SATA-I/II interfaces (1.5 to 3.0 Gbps) for recording to data drives. The SpaceCube 1.5 also features DDR2 SDRAM (double data rate synchronous dynamic random access memory); 4- Gbit Flash for storing application code for the CPU, FPGA, and DSP processing elements; and a Xilinx Platform Flash XL to store FPGA configuration files or application code. The system also incorporates a 12 bit analog to digital converter with the ability to read 32 discrete analog sensor inputs. The SpaceCube 1.5 design also has a built-in accelerometer. In addition, the system has 12 receive and transmit RS- 422 interfaces for legacy support. The SpaceCube 1.5 processor card represents the first NASA Goddard design in a compact form factor featuring the Xilinx Virtex- 5. The SpaceCube 1.5 incorporates backward compatibility with the Space- Cube 1.0 form factor and stackable architecture. It also makes use of low-cost commercial parts, but is designed for operation in harsh environments.

Geist, Alessandro↗

On the scalability of the Albany/FELIX first-order Stokes approximation ice sheet solver for large-scale simulations of the Greenland and Antarctic ice sheets

We examine the scalability of the recently developed Albany/FELIX finite-element based code for the first-order Stokes momentum balance equations for ice flow. We focus our analysis on the performance of two possible preconditioners for the iterative solution of the sparse linear systems that arise from the discretization of the governing equations: (1) a preconditioner based on the incomplete LU (ILU) factorization, and (2) a recently-developed algebraic multigrid (AMG) preconditioner, constructed using the idea of semi-coarsening. A strong scalability study on a realistic, high resolution Greenland ice sheet problem reveals that, for a given number of processor cores, the AMG preconditioner results in faster linear solve times but the ILU preconditioner exhibits better scalability. A weak scalability study is performed on a realistic, moderate resolution Antarctic ice sheet problem, a substantial fraction of which contains floating ice shelves, making it fundamentally different from the Greenland ice sheet problem. Here, we show that as the problem size increases, the performance of the ILU preconditioner deteriorates whereas the AMG preconditioner maintains scalability. This is because the linear systems are extremely ill-conditioned in the presence of floating ice shelves, and the ill-conditioning has a greater negative effect on the ILU preconditioner than on the AMG preconditioner.

54 ENVIRONMENTAL SCIENCES↗

Low-loss interconnects for modular superconducting quantum processors

Low-loss superconducting aluminium cables and on-chip impedance transformers can be used to link qubit modules and create superconducting quantum computing networks with high-fidelity intermodule state transfer. Scaling is now a key challenge in superconducting quantum computing. One solution is to build modular systems in which smaller-scale quantum modules are individually constructed and calibrated and then assembled into a larger architecture. This, however, requires the development of suitable interconnects. Here we report low-loss interconnects based on pure aluminium coaxial cables and on-chip impedance transformers featuring quality factors of up to 8.1 x 10 5 , which is comparable with the performance of our transmon qubits fabricated on a single-crystal sapphire substrate. We use these interconnects to link five quantum modules with intermodule quantum state transfer and Bell state fidelities of up to 99%. To benchmark the overall performance of the processor, we create maximally entangled, multiqubit Greenberger-Horne-Zeilinger states. The generated intermodule four-qubit Greenberger-Horne-Zeilinger state exhibits 92.0% fidelity. We also entangle up to 12 qubits in a Greenberger-Horne-Zeilinger state with 55.8 ± 1.8% fidelity, which is above the genuine multipartite entanglement threshold of 1/2.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

The SEASAT-A synthetic aperture radar design and implementation

The SEASAT-A synthetic aperture imaging radar system is the first imaging radar system intended to be used as a scientific instrument designed for orbital use. The requirement of the radar system is to generate continuous radar imagery with a 100 kilometer swath with 25 meter resolution from an orbital altitude of 800 kilometers. These requirements impose unique system design problems and a description of the implementation is given. The end-to-end system is described, including interactions of the spacecraft, antenna, sensor, telemetry link, recording subsystem, and data processor. Some of the factors leading to the selection of critical system parameters are listed. The expected error sources leading to degradation of image quality are reported as well as estimate given of the expected performance from data obtained during a ground testing of the completed subsystems.

Jordan, R. L.↗