Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel 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 991 records · Page 55

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↗

Unobtrusive Software and System Health Management with R2U2 on a Parallel MIMD Coprocessor

Dynamic monitoring of software and system health of a complex cyber-physical system requires observers that continuously monitor variables of the embedded software in order to detect anomalies and reason about root causes. There exists a variety of techniques for code instrumentation, but instrumentation might change runtime behavior and could require costly software re-certification. In this paper, we present R2U2E, a novel realization of our real-time, Realizable, Responsive, and Unobtrusive Unit (R2U2). The R2U2E observers are executed in parallel on a dedicated 16-core EPIPHANY co-processor, thereby avoiding additional computational overhead to the system under observation. A DMA-based shared memory access architecture allows R2U2E to operate without any code instrumentation or program interference.

Schumann, Johann↗

SimNet: Accurate and High-Performance Computer Architecture Simulation using Deep Learning

While cycle-accurate simulators are essential tools for architecture research, design, and development, their practicality is limited by an extremely long time-to-solution for realistic applications under investigation. This work describes a concerted effort, where machine learning (ML) is used to accelerate microarchitecture simulation. First, an ML-based instruction latency prediction framework that accounts for both static instruction properties and dynamic processor states is constructed. Then, a GPU-accelerated parallel simulator is implemented based on the proposed instruction latency predictor, and its simulation accuracy and throughput are validated and evaluated against a state-of-the-art simulator. Leveraging modern GPUs, the ML-based simulator outperforms traditional CPU-based simulators significantly.

97 MATHEMATICS AND COMPUTING↗

Implementation of Accelerated Beam-Specific Matched-Filter-Based Optical Alignment

Accurate automated alignment of laser beams in the National Ignition Facility (NIF) is essential for achieving extreme temperature and pressure required for inertial confinement fusion. The alignment achieved by the integrated control systems relies on algorithms processing video images to determine the position of the laser beam images in real-time. Alignment images that exhibit wide variations in beam quality require a matched-filter algorithm for position detection. One challenge in designing a matched-filter based algorithm is to construct a filter template that is resilient to variations in imaging conditions while guaranteeing accurate position determination. A second challenge is to process the image as fast as possible. This paper describes the development of a new analytical template that captures key recurring features present in the beam image to accurately estimate the beam position under good image quality conditions. Depending on the features present in a particular beam, the analytical template allows us to create a highly tailored template containing only those selected features. The second objective is achieved by exploiting the parallelism inherent in the algorithm to accelerate processing using parallel hardware that provides significant performance improvement over conventional processors. In particular, a Xilinx Virtex II Pro FPGA hardware implementation processing 32 templates provided a speed increase of about 253 times over an optimized software implementation running on a 2.0 GHz AMD Opteron core.

42 ENGINEERING↗

HPC Node Performance and Energy Modeling with the Co-Location of Applications

Multicore processors have become an integral part of modern large-scale and high-performance parallel and distributed computing systems. This advance became necessary as the demand for increased system performance has exceeded the limits that single core processors can provide. Unfortunately, applications co-located on multicore processors can suffer from decreased performance and increased dynamic energy use as a result of access to shared resources, such as memory. Consequently, it is increasingly important to characterize the performance of applications that execute on these architectures. This work investigates some of the disadvantages of co-location, and presents a methodology for building models capable of utilizing varying amounts of information about a target application and its co-located applications to make predictions about the target application’s execution time and the system’s energy use under arbitrary co-locations of a wide range of application types. The proposed methodology is validated on three different server class Intel Xeon multicore processors using eleven applications from two scientific benchmark suites. The model’s utility for scheduling is also demonstrated in a simulated large-scale high-performance computing environment through the creation of a co-location aware scheduling heuristic. This heuristic demonstrates that scheduling using information generated with the proposed modeling methodology is capable of making significant improvements over a scheduling heuristic that is naive to co-location interference.

performance modeling, energy modeling, resource ma↗

PFLOTRAN: Reactive Flow & Transport Code for Use on Laptops to Leadership-Class Supercomputers

PFLOTRAN, a next-generation reactive flow and transport code for modeling subsurface processes, has been designed from the ground up to run efficiently on machines ranging from leadership-class supercomputers to laptops. Based on an object-oriented design, the code is easily extensible to incorporate additional processes. It can interface seamlessly with Fortran 9X, C and C++ codes. Domain decomposition parallelism is employed, with the PETSc parallel framework used to manage parallel solvers, data structures and communication. Features of the code include a modular input file, implementation of high-performance I/O using parallel HDF5, ability to perform multiple realization simulations with multiple processors per realization in a seamless manner, and multiple modes for multiphase flow and multicomponent geochemical transport. Chemical reactions currently implemented in the code include homogeneous aqueous complexing reactions and heterogeneous mineral precipitation/dissolution, ion exchange, surface complexation and a multirate kinetic sorption model. PFLOTRAN has demonstrated petascale performance using 2{sup 17} processor cores with over 2 billion degrees of freedom. Accomplishments achieved to date include applications to the Hanford 300 Area and modeling CO{sub 2} sequestration in deep geologic formations.

58 GEOSCIENCES↗

Efficient Machine Learning Approach for Optimizing Scientific Computing Applications on Emerging HPC Architectures

Efficient parallel implementations of scientific applications on multi-core CPUs with accelerators such as GPUs and Xeon Phis is challenging. This requires - exploiting the data parallel architecture of the accelerator along with the vector pipelines of modern x86 CPU architectures, load balancing, and efficient memory transfer between different devices. It is relatively easy to meet these requirements for highly structured scientific applications. In contrast, a number of scientific and engineering applications are unstructured. Getting performance on accelerators for these applications is extremely challenging because many of these applications employ irregular algorithms which exhibit data-dependent control-ow and irregular memory accesses. Furthermore, these applications are often iterative with dependency between steps, and thus making it hard to parallelize across steps. As a result, parallelism in these applications is often limited to a single step. Numerical simulation of charged particles beam dynamics is one such application where the distribution of work and memory access pattern at each time step is irregular. Applications with these properties tend to present significant branch and memory divergence, load imbalance between different processor cores, and poor compute and memory utilization. Prior research on parallelizing such irregular applications have been focused around optimizing the irregular, data-dependent memory accesses and control-ow during a single step of the application independent of the other steps, with the assumption that these patterns are completely unpredictable. We observed that the structure of computation leading to control-ow divergence and irregular memory accesses in one step is similar to that in the next step. It is possible to predict this structure in the current step by observing the computation structure of previous steps. In this dissertation, we present novel machine learning based optimization techniques to address the parallel implementation challenges of such irregular applications on different HPC architectures. In particular, we use supervised learning to predict the computation structure and use it to address the control-ow and memory access irregularities in the parallel implementation of such applications on GPUs, Xeon Phis, and heterogeneous architectures composed of multi-core CPUs with GPUs or Xeon Phis. We use numerical simulation of charged particles beam dynamics simulation as a motivating example throughout the dissertation to present our new approach, though they should be equally applicable to a wide range of irregular applications. The machine learning approach presented here use predictive analytics and forecasting techniques to adaptively model and track the irregular memory access pattern at each time step of the simulation to anticipate the future memory access pattern. Access pattern forecasts can then be used to formulate optimization decisions during application execution which improves the performance of the application at a future time step based on the observations from earlier time steps. In heterogeneous architectures, forecasts can also be used to improve the memory performance and resource utilization of all the processing units to deliver a good aggregate performance. We used these optimization techniques and anticipation strategy to design a cache-aware, memory efficient parallel algorithm to address the irregularities in the parallel implementation of charged particles beam dynamics simulation on different HPC architectures. Experimental result using a diverse mix of HPC architectures shows that our approach in using anticipation strategy is effective in maximizing data reuse, ensuring workload balance, minimizing branch and memory divergence, and in improving resource utilization.

Arumugam, Kamesh↗

Parallel processing of a rotating shaft simulation

A FORTRAN program describing the vibration modes of a rotor-bearing system is analyzed for parellelism in this simulation using a Pascal-like structured language. Potential vector operations are also identified. A critical path through the simulation is identified and used in conjunction with somewhat fictitious processor characteristics to determine the time to calculate the problem on a parallel processing system having those characteristics. A parallel processing overhead time is included as a parameter for proper evaluation of the gain over serial calculation. The serial calculation time is determined for the same fictitious system. An improvement of up to 640 percent is possible depending on the value of the overhead time. Based on the analysis, certain conclusions are drawn pertaining to the development needs of parallel processing technology, and to the specification of parallel processing systems to meet computational needs.

Arpasi, Dale J.↗

Streamline Integration Using MPI-Hybrid Parallelism on a Large Multicore Architecture

Streamline computation in a very large vector field data set represents a significant challenge due to the nonlocal and data-dependent nature of streamline integration. In this paper, we conduct a study of the performance characteristics of hybrid parallel programming and execution as applied to streamline integration on a large, multicore platform. With multicore processors now prevalent in clusters and supercomputers, there is a need to understand the impact of these hybrid systems in order to make the best implementation choice. We use two MPI-based distribution approaches based on established parallelization paradigms, parallelize over seeds and parallelize over blocks, and present a novel MPI-hybrid algorithm for each approach to compute streamlines. Our findings indicate that the work sharing between cores in the proposed MPI-hybrid parallel implementation results in much improved performance and consumes less communication and I/O bandwidth than a traditional, nonhybrid distributed implementation.

97 MATHEMATICS AND COMPUTING↗

Visualization Co-Processing of a CFD Simulation

OVERFLOW, a widely used CFD simulation code, is combined with a visualization system, pV3, to experiment with an environment for simulation/visualization co-processing on a SGI Origin 2000 computer(O2K) system. The shared memory version of the solver is used with the O2K 'pfa' preprocessor invoked to automatically discover parallelism in the source code. No other explicit parallelism is enabled. In order to study the scaling and performance of the visualization co-processing system, sample runs are made with different processor groups in the range of 1 to 254 processors. The data exchange between the visualization system and the simulation system is rapid enough for user interactivity when the problem size is small. This shared memory version of OVERFLOW, with minimal parallelization, does not scale well to an increasing number of available processors. The visualization task takes about 18 to 30% of the total processing time and does not appear to be a major contributor to the poor scaling. Improper load balancing and inter-processor communication overhead are contributors to this poor performance. Work is in progress which is aimed at obtaining improved parallel performance of the solver and removing the limitations of serial data transfer to pV3 by examining various parallelization/communication strategies, including the use of the explicit message passing.

Vaziri, Arsi↗

Spacecraft on-board SAR processing technology

This paper provides an assessment of the on-board SAR processing technology for Eos-type missions. The proposed Eos SAR sensor and flight data system are introduced, and the SAR processing requirements are described. The SAR on-board SAR processor architecture selection is discussed, and a baseline processor architecture using a frequency-domain processor for range correlation and a modular fault-tolerant VLSI time-domain parallel array for azimuth correlation are described. The mass storage and VLSI technologies needed for implementing the proposed SAR processing are assessed. It is shown that acceptable processor power and mass characteristics should be feasible for Eos-type applications. A proposed development strategy for the on-board SAR processor is presented.

Liu, K. Y.↗

An Analog Processor for Image Compression

This paper describes a novel analog Vector Array Processor (VAP) that was designed for use in real-time and ultra-low power image compression applications. This custom CMOS processor is based architectually on the Vector Quantization (VQ) algorithm in image coding, and the hardware implementation fully exploits the inherent parallelism built-in the VQ algorithm.

Vector Array Processor VAP analog processor image ↗

Error estimation and adaptive mesh refinement for parallel analysis of shell structures

The formulation and application of element-level, element-independent error indicators is investigated. This research culminates in the development of an error indicator formulation which is derived based on the projection of element deformation onto the intrinsic element displacement modes. The qualifier 'element-level' means that no information from adjacent elements is used for error estimation. This property is ideally suited for obtaining error values and driving adaptive mesh refinements on parallel computers where access to neighboring elements residing on different processors may incur significant overhead. In addition such estimators are insensitive to the presence of physical interfaces and junctures. An error indicator qualifies as 'element-independent' when only visible quantities such as element stiffness and nodal displacements are used to quantify error. Error evaluation at the element level and element independence for the error indicator are highly desired properties for computing error in production-level finite element codes. Four element-level error indicators have been constructed. Two of the indicators are based on variational formulation of the element stiffness and are element-dependent. Their derivations are retained for developmental purposes. The second two indicators mimic and exceed the first two in performance but require no special formulation of the element stiffness mesh refinement which we demonstrate for two dimensional plane stress problems. The parallelizing of substructures and adaptive mesh refinement is discussed and the final error indicator using two-dimensional plane-stress and three-dimensional shell problems is demonstrated.

Keating, Scott C.↗

Performance prediction: A case study using a multi-ring KSR-1 machine

While computers with tens of thousands of processors have successfully delivered high performance power for solving some of the so-called 'grand-challenge' applications, the notion of scalability is becoming an important metric in the evaluation of parallel machine architectures and algorithms. In this study, the prediction of scalability and its application are carefully investigated. A simple formula is presented to show the relation between scalability, single processor computing power, and degradation of parallelism. A case study is conducted on a multi-ring KSR1 shared virtual memory machine. Experimental and theoretical results show that the influence of topology variation of an architecture is predictable. Therefore, the performance of an algorithm on a sophisticated, heirarchical architecture can be predicted and the best algorithm-machine combination can be selected for a given application.

Sun, Xian-He↗

Active Flash: Performance-Energy Tradeoffs for Out-of-Core Processing on Non-Volatile Memory Devices

In this abstract, we study the performance and energy tradeoffs involved in migrating data analysis into the flash device, a process we refer to as Active Flash. The Active Flash paradigm is similar to 'active disks', which has received considerable attention. Active Flash allows us to move processing closer to data, thereby minimizing data movement costs and reducing power consumption. It enables true out-of-core computation. The conventional definition of out-of-core solvers refers to an approach to process data that is too large to fit in the main memory and, consequently, requires access to disk. However, in Active Flash, processing outside the host CPU literally frees the core and achieves real 'out-of-core' analysis. Moving analysis to data has long been desirable, not just at this level, but at all levels of the system hierarchy. However, this requires a detailed study on the tradeoffs involved in achieving analysis turnaround under an acceptable energy envelope. To this end, we first need to evaluate if there is enough computing power on the flash device to warrant such an exploration. Flash processors require decent computing power to run the internal logic pertaining to the Flash Translation Layer (FTL), which is responsible for operations such as address translation, garbage collection (GC) and wear-leveling. Modern SSDs are composed of multiple packages and several flash chips within a package. The packages are connected using multiple I/O channels to offer high I/O bandwidth. SSD computing power is also expected to be high enough to exploit such inherent internal parallelism within the drive to increase the bandwidth and to handle fast I/O requests. More recently, SSD devices are being equipped with powerful processing units and are even embedded with multicore CPUs (e.g. ARM Cortex-A9 embedded processor is advertised to reach 2GHz frequency and deliver 5000 DMIPS; OCZ RevoDrive X2 SSD has 4 SandForce controllers, each with 780MHz max frequency Tensilica core). Efforts that take advantage of the available computing cycles on the processors on SSDs to run auxiliary tasks other than actual I/O requests are beginning to emerge. Kim et al. investigate database scan operations in the context of processing on the SSDs, and propose dedicated hardware logic to speed up scans. Also, cluster architectures have been explored, which consist of low-power embedded CPUs coupled with small local flash to achieve fast, parallel access to data. Processor utilization on SSD is highly dependent on workloads and, therefore, they can be idle during periods with no I/O accesses. We propose to use the available processing capability on the SSD to run tasks that can be offloaded from the host. This paper makes the following contributions: (1) We have investigated Active Flash and its potential to optimize the total energy cost, including power consumption on the host and the flash device; (2) We have developed analytical models to analyze the performance-energy tradeoffs for Active Flash, by treating the SSD as a blackbox, this is particularly valuable due to the proprietary nature of the SSD internal hardware; and (3) We have enhanced a well-known SSD simulator (from MSR) to implement 'on-the-fly' data compression using Active Flash. Our results provide a window into striking a balance between energy consumption and application performance.

42 ENGINEERING↗

A PARALLEL MONTE CARLO CODE FOR SIMULATING COLLISIONAL N-BODY SYSTEMS

We present a new parallel code for computing the dynamical evolution of collisional N-body systems with up to N {approx} 10{sup 7} particles. Our code is based on the Henon Monte Carlo method for solving the Fokker-Planck equation, and makes assumptions of spherical symmetry and dynamical equilibrium. The principal algorithmic developments involve optimizing data structures and the introduction of a parallel random number generation scheme as well as a parallel sorting algorithm required to find nearest neighbors for interactions and to compute the gravitational potential. The new algorithms we introduce along with our choice of decomposition scheme minimize communication costs and ensure optimal distribution of data and workload among the processing units. Our implementation uses the Message Passing Interface library for communication, which makes it portable to many different supercomputing architectures. We validate the code by calculating the evolution of clusters with initial Plummer distribution functions up to core collapse with the number of stars, N, spanning three orders of magnitude from 10{sup 5} to 10{sup 7}. We find that our results are in good agreement with self-similar core-collapse solutions, and the core-collapse times generally agree with expectations from the literature. Also, we observe good total energy conservation, within {approx}< 0.04% throughout all simulations. We analyze the performance of the code, and demonstrate near-linear scaling of the runtime with the number of processors up to 64 processors for N = 10{sup 5}, 128 for N = 10{sup 6} and 256 for N = 10{sup 7}. The runtime reaches saturation with the addition of processors beyond these limits, which is a characteristic of the parallel sorting algorithm. The resulting maximum speedups we achieve are approximately 60 Multiplication-Sign , 100 Multiplication-Sign , and 220 Multiplication-Sign , respectively.

79 ASTRONOMY AND ASTROPHYSICS↗

Optical systolic array processor using residue arithmetic

The use of residue arithmetic to increase the accuracy and reduce the dynamic range requirements of optical matrix-vector processors is evaluated. It is determined that matrix-vector operations and iterative algorithms can be performed totally in residue notation. A new parallel residue quantizer circuit is developed which significantly improves the performance of the systolic array feedback processor. Results are presented of a computer simulation of this system used to solve a set of three simultaneous equations.

Jackson, J.↗

Parallelized reliability estimation of reconfigurable computer networks

A parallelized system, ASSURE, for computing the reliability of embedded avionics flight control systems which are able to reconfigure themselves in the event of failure is described. ASSURE accepts a grammar that describes a reliability semi-Markov state-space. From this it creates a parallel program that simultaneously generates and analyzes the state-space, placing upper and lower bounds on the probability of system failure. ASSURE is implemented on a 32-node Intel iPSC/860, and has achieved high processor efficiencies on real problems. Through a combination of improved algorithms, exploitation of parallelism, and use of an advanced microprocessor architecture, ASSURE has reduced the execution time on substantial problems by a factor of one thousand over previous workstation implementations. Furthermore, ASSURE's parallel execution rate on the iPSC/860 is an order of magnitude faster than its serial execution rate on a Cray-2 supercomputer. While dynamic load balancing is necessary for ASSURE's good performance, it is needed only infrequently; the particular method of load balancing used does not substantially affect performance.

Nicol, David M.↗