Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel systems”

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 1,351 records · Page 75

Enabling event tracing at leadership-class scale through I/O forwarding middleware. In: HPDC '12 Proceedings of the 21st international symposium on High-Performance Parallel and Distributed Computing

Event tracing is an important tool for understanding the performance of parallel applications. As concurrency increases in leadership-class computing systems, the quantity of performance log data can overload the parallel file system, perturbing the application being observed. In this work we present a solution for event tracing at leadership scales. We enhance the I/O forwarding system software to aggregate and reorganize log data prior to writing to the storage system, significantly reducing the burden on the underlying file system for this type of traffic. Furthermore, we augment the I/O forwarding system with a write buffering capability to limit the impact of artificial perturbations from log data accesses on traced applications. To validate the approach, we modify the Vampir tracing toolset to take advantage of this new capability and show that the approach increases the maximum traced application size by a factor of 5x to more than 200,000 processes.

Ilsche, Thomas↗

An Efficient Implementation of Multiscale Simulation Software PNP-cDFT

An efficient implementation of PNP-cDFT, a multiscale method for computing the chemical potentials of charged species is designed and evaluated. Spatial decomposition of the multi particle system is employed in the parallelization of classical density functional theory (cDFT) algorithm. Furthermore, a truncated strategy is used to reduce the computational complexity of cDFT algorithm. The simulation results show that the parallel implementation has close to linear scalability in parallel computing environments for both 1D and 3D systems. It also shows that the truncated versions of cDFT improve the efficiency of the methods substantially.

99 GENERAL AND MISCELLANEOUS↗

Enabling Event Tracing at Leadership-Class Scale through I/O Forwarding Middleware

Event tracing is an important tool for understanding the performance of parallel applications. As concurrency increases in leadership-class computing systems, the quantity of performance log data can overload the parallel file system, perturbing the application being observed. In this work we present a solution for event tracing at leadership scales. We enhance the I/O forwarding system software to aggregate and reorganize log data prior to writing to the storage system, significantly reducing the burden on the underlying file system for this type of traffic. Furthermore, we augment the I/O forwarding system with a write buffering capability to limit the impact of artificial perturbations from log data accesses on traced applications. To validate the approach, we modify the Vampir tracing tool to take advantage of this new capability and show that the approach increases the maximum traced application size by a factor of 5x to more than 200,000 processors.

Ilsche, Thomas↗

A Parallel Trade Study Architecture for Design Optimization of Complex Systems

Design of a successful product requires evaluating many design alternatives in a limited design cycle time. This can be achieved through leveraging design space exploration tools and available computing resources on the network. This paper presents a parallel trade study architecture to integrate trade study clients and computing resources on a network using Web services. The parallel trade study solution is demonstrated to accelerate design of experiments, genetic algorithm optimization, and a cost as an independent variable (CAIV) study for a space system application.

Kim, Hongman↗

An experimental and theoretical study of adenine adsorption on Au(111)

A model study of adenine adsorption on the Au(111) surface is reported for molecular adlayers prepared by evaporation in vacuum and deposition from saturated aqueous solution. The electronic structure and adsorption geometry of the molecular films were studied experimentally by X-ray photoelectron spectroscopy and near edge X-ray absorption fine structure spectroscopy. Adsorption models are proposed for the adlayers arising from the different preparation methods. Density functional theory calculations were used to examine both parallel and upright adenine adsorption geometries, supply additional information on the bond strength, and identify which atom is involved in bonding to Au(111). In the case of deposition in vacuum, the adenine molecule is bound via van der Waals forces to Au(111) with the molecular plane parallel to the surface, consistent with the published scanning tunneling microscopy data on this system. The most stable parallel adenine configuration was found to have an adsorption energy of ca. -1.1 eV using the optB86b-vdW functional. For adenine deposition from aqueous solution, the adlayer is disordered, with molecules in an upright geometry, and with an adsorption energy of ca. -1.0 eV, coordinated via the imino N3 nitrogen atom. The present study contributes to the substantial literature of model studies of adenine on Au(111), complementing the existing knowledge with information on electronic structure, bonding geometry and adsorption energy of this system.

Acres, Robert G.↗

Spatial interferometry for white light processing: The coherence interferometer

An optical systems design approach for a parallel optical processor that has an information throughput capacity in excess of one petabit per second is discussed. This system enables predetection processing of white light, passively illuminated scenes. Integrated with a programmable Van der Lugt filter, this system is an effective data compression method. A program to implement this parallel optical processing by interferometry system is discussed.

Breckinridge, J. B.↗

Anderson acceleration of the Jacobi iterative method: An efficient alternative to Krylov methods for large, sparse linear systems

We employ Anderson extrapolation to accelerate the classical Jacobi iterative method for large, sparse linear systems. Specifically, we utilize extrapolation at periodic intervals within the Jacobi iteration to develop the Alternating Anderson–Jacobi (AAJ) method. We verify the accuracy and efficacy of AAJ in a range of test cases, including nonsymmetric systems of equations. We demonstrate that AAJ possesses a favorable scaling with system size that is accompanied by a small prefactor, even in the absence of a preconditioner. In particular, we show that AAJ is able to accelerate the classical Jacobi iteration by over four orders of magnitude, with speed-ups that increase as the system gets larger. Moreover, we find that AAJ significantly outperforms the Generalized Minimal Residual (GMRES) method in the range of problems considered here, with the relative performance again improving with size of the system. As a result, the proposed method represents a simple yet efficient technique that is particularly attractive for large-scale parallel solutions of linear systems of equations.

97 MATHEMATICS AND COMPUTING↗

GPU-Accelerated Solution of the Bethe–Salpeter Equation for Large and Heterogeneous Systems

We present a massively parallel GPU-accelerated implementation of the Bethe–Salpeter equation (BSE) for the calculation of the vertical excitation energies (VEEs) and optical absorption spectra of condensed and molecular systems, starting from single-particle eigenvalues and eigenvectors obtained with density functional theory. The algorithms adopted here circumvent the slowly converging sums over empty and occupied states and the inversion of large dielectric matrices through a density matrix perturbation theory approach and a low-rank decomposition of the screened Coulomb interaction, respectively. Further computational savings are achieved by exploiting the nearsightedness of the density matrix of semiconductors and insulators to reduce the number of screened Coulomb integrals. We scale our calculations to thousands of GPUs with a hierarchical loop and data distribution strategy. The efficacy of our method is demonstrated by computing the VEEs of several spin defects in wide-band-gap materials, showing that supercells with up to 1000 atoms are necessary to obtain converged results. We discuss the validity of the common approximation that solves the BSE with truncated sums over empty and occupied states. In conclusion, we then apply our GW-BSE implementation to a diamond lattice with 1727 atoms to study the symmetry breaking of triplet states caused by the interaction of a point defect with an extended line defect.

Absorption spectra↗

Prometheus: Scalable and Accurate Emulation of Task-Based Applications on Many-Core Systems.

Modeling the performance of non-deterministic parallel applications on future many-core systems requires the development of novel simulation and emulation techniques and tools. We present “Prometheus”, a fast, accurate and modular emulation framework for task-based applications. By raising the level of abstraction and focusing on runtime synchronization, Prometheus can accurately predict applications’ performance on very large many-core systems. We validate our emulation framework against two real platforms (AMD Interlagos and Intel MIC) and report error rates generally below 4%. We, then, evaluate Prometheus’ performance and scalability: our results show that Prometheus can emulate a task-based application on a system with 512K cores in 11.5 hours. We present two test cases that show how Prometheus can be used to study the performance and behavior of systems that present some of the characteristics expected from exascale supercomputer nodes, such as active power management and processors with a high number of cores but reduced cache per core.

exascale systems↗

Real-time data system: Incorporating new technology in mission critical environments

If the Space Station Freedom is to remain viable over its 30-year life span, it must be able to incorporate new information systems technologies. These technologies are necessary to enhance mission effectiveness and to enable new NASA missions, such as supporting the Lunar-Mars Initiative. Hi-definition television (HDTV), neural nets, model-based reasoning, advanced languages, CPU designs, and computer networking standards are areas which have been forecasted to make major strides in the next 30 years. A major challenge to NASA is to bring these technologies online without compromising mission safety. In past programs, NASA managers have been understandably reluctant to rely on new technologies for mission critical activities until they are proven in noncritical areas. NASA must develop strategies to allow inflight confidence building and migration of technologies into the trusted tool base. NASA has successfully met this challenge and developed a winning strategy in the Space Shuttle Mission Control Center. This facility, which is clearly among NASA's most critical, is based on 1970's mainframe architecture. Changes to the mainframe are very expensive due to the extensive testing required to prove that changes do not have unanticipated impact on critical processes. Systematic improvement efforts in this facility have been delayed due to this 'risk to change.' In the real-time data system (RTDS) we have introduced a network of engineering computer workstations which run in parallel to the mainframe system. These workstations are located next to flight controller operating positions in mission control and, in some cases, the display units are mounted in the traditional mainframe consoles. This system incorporates several major improvements over the mainframe consoles including automated fault detection by real-time expert systems and color graphic animated schematics of subsystems driven by real-time telemetry. The workstations have the capability of recording telemetry data and providing 'instant replay' for flight controllers. RTDS also provides unique graphics animated by real-time telemetry such as workstation emulation of the shuttle's flight instruments and displays of the remote manipulator system (RMS) position. These systems have been used successfully as prime operational tools since STS-26 and have supported seven shuttle missions.

Muratore, John F.↗

Resource Prospector Propulsion System Cold Flow Testing

Resource Prospector (RP) is a NASA mission being led by NASA Ames Research Center with current plans to deliver a scientific payload package aboard a rover to the lunar surface. As part of an early risk reduction activity, Marshall Space Flight Center (MSFC) and Johnson Space Flight Center (JSC) have jointly developed a government-version concept of a lunar lander for the mission. The spacecraft consists of two parts, the lander and the rover which carries the scientific instruments. The lander holds the rover during launch, cruise, and landing on the surface. Following terminal descent and landing the lander portion of the spacecraft become dormant after the rover embarks on the science mission. The lander will be equipped with a propulsion system for lunar descent and landing, as well as trajectory correction and attitude control maneuvers during transit to the moon. Hypergolic propellants monomethyl hydrazine and nitrogen tetroxide will be used to fuel sixteen 70-lbf descent thrusters and twelve 5-lbf attitude control thrusters. A total of four metal-diaphragm tanks, two per propellant, will be used along with a high-pressure composite-overwrapped pressure vessel for the helium pressurant gas. Many of the major propulsion system components are heritage missile hardware obtained by NASA from the Air Force. In parallel with the flight system design activities, a simulated propulsion system based on flight drawings was built for conducting a series of water flow tests to characterize the transient fluid flow of the propulsion system feed lines and to verify the critical operation modes such as system priming, waterhammer, and crucial mission duty cycles. The primary objective of the cold flow testing was to simulate the RP propulsion system fluid flow operation through water flow testing and to obtain data for anchoring analytical models. The models will be used to predict the transient and steady state flow behaviors in the actual flight operations. All design and build efforts, including the analytical modeling, have been performed. The cold flow testing of the propulsion system was set up and conducted at a NASA MSFC test facility. All testing was completed in the summer of 2014, and this paper documents the results of that testing and the associated fluid system modeling efforts.

Williams, Hunter↗

Adaptive parallel logic networks

Adaptive, self-organizing concurrent systems (ASOCS) that combine self-organization with massive parallelism for such applications as adaptive logic devices, robotics, process control, and system malfunction management, are presently discussed. In ASOCS, an adaptive network composed of many simple computing elements operating in combinational and asynchronous fashion is used and problems are specified by presenting if-then rules to the system in the form of Boolean conjunctions. During data processing, which is a different operational phase from adaptation, the network acts as a parallel hardware circuit.

Martinez, Tony R.↗

Parallel O(log n) algorithms for open- and closed-chain rigid multibody systems based on a new mass matrix factorization technique

In this paper, parallel O(log n) algorithms for computation of rigid multibody dynamics are developed. These parallel algorithms are derived by parallelization of new O(n) algorithms for the problem. The underlying feature of these O(n) algorithms is a drastically different strategy for decomposition of interbody force which leads to a new factorization of the mass matrix (M). Specifically, it is shown that a factorization of the inverse of the mass matrix in the form of the Schur Complement is derived as M(exp -1) = C - B(exp *)A(exp -1)B, wherein matrices C, A, and B are block tridiagonal matrices. The new O(n) algorithm is then derived as a recursive implementation of this factorization of M(exp -1). For the closed-chain systems, similar factorizations and O(n) algorithms for computation of Operational Space Mass Matrix lambda and its inverse lambda(exp -1) are also derived. It is shown that these O(n) algorithms are strictly parallel, that is, they are less efficient than other algorithms for serial computation of the problem. But, to our knowledge, they are the only known algorithms that can be parallelized and that lead to both time- and processor-optimal parallel algorithms for the problem, i.e., parallel O(log n) algorithms with O(n) processors. The developed parallel algorithms, in addition to their theoretical significance, are also practical from an implementation point of view due to their simple architectural requirements.

Fijany, Amir↗

Parallel-Processing Test Bed For Simulation Software

Second-generation Hypercluster computing system is multiprocessor test bed for research on parallel algorithms for simulation in fluid dynamics, electromagnetics, chemistry, and other fields with large computational requirements but relatively low input/output requirements. Built from standard, off-shelf hardware readily upgraded as improved technology becomes available. System used for experiments with such parallel-processing concepts as message-passing algorithms, debugging software tools, and computational steering. First-generation Hypercluster system described in "Hypercluster Parallel Processor" (LEW-15283).

Blech, Richard↗

Reducing communication in algebraic multigrid with multi-step node aware communication

Algebraic multigrid (AMG) is often viewed as a scalable [Formula: see text] solver for sparse linear systems. Yet, AMG lacks parallel scalability due to increasingly large costs associated with communication, both in the initial construction of a multigrid hierarchy and in the iterative solve phase. This work introduces a parallel implementation of AMG that reduces the cost of communication, yielding improved parallel scalability. It is common in Message Passing Interface (MPI), particularly in the MPI-everywhere approach, to arrange inter-process communication, so that communication is transported regardless of the location of the send and receive processes. Performance tests show notable differences in the cost of intra- and internode communication, motivating a restructuring of communication. In this case, the communication schedule takes advantage of the less costly intra-node communication, reducing both the number and the size of internode messages. Node-centric communication extends to the range of components in both the setup and solve phase of AMG, yielding an increase in the weak and strong scaling of the entire method.

Computer Science↗

Commissioning and Testing of a new 4.5K Cold Box for JLab Cryogenic Test Facility

A new 4.5K cold box at Jefferson Lab (JLab) Cryogenic Test Facility (CTF) was recently installed and commissioned to upgrade the existing 4.5 K refrigeration system and work in parallel with the existing 4.5 K cold box. This new 4.5 K cold box is equipped with two turbo-expanders and, at its maximum capacity condition, it is expected to support at least 6.6 g/s of helium liquefaction or 650 W of refrigeration at 4.5 K. It can also handle up to 10 g/s 30 K return flow for 2 K refrigeration recovery that supports cryo-module and superconducting cavity testing. Performances of the cold box at its maximum capacity conditions as well as several other operating modes were tested for acceptance. Finally, we will briefly review the new 4.5 K cold box design features and discuss the commissioning and performance testing results.

42 ENGINEERING↗

Rocket Engine Numerical Simulator (RENS)

Work is being done at three universities to help today's NASA engineers use the knowledge and experience of their Apolloera predecessors in designing liquid rocket engines. Ground-breaking work is being done in important subject areas to create a prototype of the most important functions for the Rocket Engine Numerical Simulator (RENS). The goal of RENS is to develop an interactive, realtime application that engineers can utilize for comprehensive preliminary propulsion system design functions. RENS will employ computer science and artificial intelligence research in knowledge acquisition, computer code parallelization and objectification, expert system architecture design, and object-oriented programming. In 1995, a 3year grant from the NASA Lewis Research Center was awarded to Dr. Douglas Moreman and Dr. John Dyer of Southern University at Baton Rouge, Louisiana, to begin acquiring knowledge in liquid rocket propulsion systems. Resources of the University of West Florida in Pensacola were enlisted to begin the process of enlisting knowledge from senior NASA engineers who are recognized experts in liquid rocket engine propulsion systems. Dr. John Coffey of the University of West Florida is utilizing his expertise in interviewing and concept mapping techniques to encode, classify, and integrate information obtained through personal interviews. The expertise extracted from the NASA engineers has been put into concept maps with supporting textual, audio, graphic, and video material. A fundamental concept map was delivered by the end of the first year of work and the development of maps containing increasing amounts of information is continuing. Find out more information about this work at the Southern University/University of West Florida. In 1996, the Southern University/University of West Florida team conducted a 4day group interview with a panel of five experts to discuss failures of the RL10 rocket engine in conjunction with the Centaur launch vehicle. The discussion was recorded on video and audio tape. Transcriptions of the entire proceedings and an abbreviated video presentation of the discussion highlights are under development. Also in 1996, two additional 3year grants were awarded to conduct parallel efforts that would complement the work being done by Southern University and the University of West Florida. Dr. Prem Bhalla of Jackson State University in Jackson, Mississippi, is developing the architectural framework for RENS. By employing the Rose Rational language and Booch Object Oriented Programming (OOP) technology, Dr. Bhalla is developing the basic structure of RENS by identifying and encoding propulsion system components, their individual characteristics, and cross-functionality and dependencies. Dr. Ruknet Cezzar of Hampton University, located in Hampton, Virginia, began working on the parallelization and objectification of rocket engine analysis and design codes. Dr. Cezzar will use the Turbo C++ OOP language to translate important liquid rocket engine computer codes from FORTRAN and permit their inclusion into the RENS framework being developed at Jackson State University. The Southern University/University of West Florida grant was extended by 1 year to coordinate the conclusion of all three efforts in 1999.

Davidian, Kenneth O.↗