Search NASA⌕ Search

SEARCH · Search NASA

Results for “Threading”

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 613 records · Page 34

Implementation of NAS Parallel Benchmarks in Java

A number of features make Java an attractive but a debatable choice for High Performance Computing (HPC). In order to gauge the applicability of Java to the Computational Fluid Dynamics (CFD) we have implemented NAS Parallel Benchmarks in Java. The performance and scalability of the benchmarks point out the areas where improvement in Java compiler technology and in Java thread implementation would move Java closer to Fortran in the competition for CFD applications.

Frumkin, Michael↗

Enabling Advanced Automation in Spacecraft Operations with the Spacecraft Emergency Response System

True autonomy is the Holy Grail of spacecraft mission operations. The goal of launching a satellite and letting it manage itself throughout its useful life is a worthy one. With true autonomy, the cost of mission operations would be reduced to a negligible amount. Under full autonomy, any problems (no matter the severity or type) that may arise with the spacecraft would be handled without any human intervention via some combination of smart sensors, on-board intelligence, and/or smart automated ground system. Until the day that complete autonomy is practical and affordable to deploy, incremental steps of deploying ever-increasing levels of automation (computerization of once manual tasks) on the ground and on the spacecraft are gradually decreasing the cost of mission operations. For example, NASA's Goddard Space Flight Center (NASA-GSFC) has been flying spacecraft with low cost operations for several years. NASA-GSFC's SMEX (Small Explorer) and MIDEX (Middle Explorer) missions have effectively deployed significant amounts of automation to enable the missions to fly predominately in 'light-out' mode. Under light-out operations the ground system is run without human intervention. Various tools perform many of the tasks previously performed by the human operators. One of the major issues in reducing human staff in favor of automation is the perceived increased in risk of losing data, or even losing a spacecraft, because of anomalous conditions that may occur when there is no one in the control center. When things go wrong, missions deploying advanced automation need to be sure that anomalous conditions are detected and that key personal are notified in a timely manner so that on-call team members can react to those conditions. To ensure the health and safety of its lights-out missions, NASA-GSFC's Advanced Automation and Autonomy branch (Code 588) developed the Spacecraft Emergency Response System (SERS). The SERS is a Web-based collaborative environment that enables secure distributed fault and resource management. The SERS incorporates the use of intelligent agents, threaded discussions, workflow, database connectivity, and links to a variety of communications devices (e.g., two-way paging, PDA's, and Internet phones) via commercial gateways. When the SERS detects a problem, it notifies on-call team members, who then can remotely take any necessary actions to resolve the anomalies.The SERS goes well beyond a simple '911' system that sends out an error code to everyone with a pager. Instead, SERS' software agents send detailed data (i.e., notifications) to the most appropriate team members based on the type and severity of the anomaly and the skills of the on-call team members. The SERS also allows the team members to respond to the notifications from their wireless devices. This unique capability ensures rapid response since the team members no longer have to go to a PC or the control center for every anomalous event. Most importantly, the SERS enables safe experimentation with various techniques for increasing levels of automation, leading to robust autonomy. For the MIDEX missions at NASA GSFC, the SERS is used to provide 'human-in-the-loop' automation. During lights-out operations, as greater control is given to the MIDEX automated systems, the SERS can be configured to page remote personnel and keep them informed regarding actions taking place in the control center. Remote off-duty operators can even be given the option of enabling or inhibiting a specific automated response in near real time via their two-way pagers. The SERS facilitates insertion of new technology to increase automation, while maintaining the safety and security of mission resources. This paper will focus on SERS' overall functionality and how SERS has been designed to handle the monitoring and emergency response for missions with varying levels of automation. The paper will also convey some of the key lessons learned from SERS' deployment across of variety of missions, highlighting this incremental approach to achieving 'robust autonomy'.

Breed, Julie↗

Accelerating Demand Paging for Local and Remote Out-of-Core Visualization

This paper describes a new algorithm that improves the performance of application-controlled demand paging for the out-of-core visualization of data sets that are on either local disks or disks on remote servers. The performance improvements come from better overlapping the computation with the page reading process, and by performing multiple page reads in parallel. The new algorithm can be applied to many different visualization algorithms since application-controlled demand paging is not specific to any visualization algorithm. The paper includes measurements that show that the new multi-threaded paging algorithm decreases the time needed to compute visualizations by one third when using one processor and reading data from local disk. The time needed when using one processor and reading data from remote disk decreased by up to 60%. Visualization runs using data from remote disk ran about as fast as ones using data from local disk because the remote runs were able to make use of the remote server's high performance disk array.

Ellsworth, David↗

Pre-Hardware Optimization of Spacecraft Image Processing Software Algorithms and Hardware Implementation

Spacecraft telemetry rates have steadily increased over the last decade presenting a problem for real-time processing by ground facilities. This paper proposes a solution to a related problem for the Geostationary Operational Environmental Spacecraft (GOES-8) image processing application. Although large super-computer facilities are the obvious heritage solution, they are very costly, making it imperative to seek a feasible alternative engineering solution at a fraction of the cost. The solution is based on a Personal Computer (PC) platform and synergy of optimized software algorithms and re-configurable computing hardware technologies, such as Field Programmable Gate Arrays (FPGA) and Digital Signal Processing (DSP). It has been shown in [1] and [2] that this configuration can provide superior inexpensive performance for a chosen application on the ground station or on-board a spacecraft. However, since this technology is still maturing, intensive pre-hardware steps are necessary to achieve the benefits of hardware implementation. This paper describes these steps for the GOES-8 application, a software project developed using Interactive Data Language (IDL) (Trademark of Research Systems, Inc.) on a Workstation/UNIX platform. The solution involves converting the application to a PC/Windows/RC platform, selected mainly by the availability of low cost, adaptable high-speed RC hardware. In order for the hybrid system to run, the IDL software was modified to account for platform differences. It was interesting to examine the gains and losses in performance on the new platform, as well as unexpected observations before implementing hardware. After substantial pre-hardware optimization steps, the necessity of hardware implementation for bottleneck code in the PC environment became evident and solvable beginning with the methodology described in [1], [2], and implementing a novel methodology for this specific application [6]. The PC-RC interface bandwidth problem for the class of applications with moderate input-output data rates but large intermediate multi-thread data streams has been addressed and mitigated. This opens a new class of satellite image processing applications for bottleneck problems solution using RC technologies. The issue of a science algorithm level of abstraction necessary for RC hardware implementation is also described. Selected Matlab functions already implemented in hardware were investigated for their direct applicability to the GOES-8 application with the intent to create a library of Matlab and IDL RC functions for ongoing work. A complete class of spacecraft image processing applications using embedded re-configurable computing technology to meet real-time requirements, including performance results and comparison with the existing system, is described in this paper.

Kizhner, Semion↗

Metal Flow During Friction Stir Welding

The flow of metal during Friction Stir Welding is clarified using a faying surface tracer and a nib frozen in place during welding. It is shown that material is transported by two processes. The first is a wiping of material from the advancing front side of the nib onto a plug of material that rotates and advances with the nib. The material undergoes a helical motion within the plug that both rotates and advances with the plug and descends in the wash of the threads on the nib and rises on the outer part of the plug. After one or more rotations, this material is sloughed off the plug in its wake, primarily on the advancing side. The second process is an entrainment of material from the front retreating side of the nib that fills in between the sloughed off pieces from the advancing side. These two processes produce material with different mechanical properties and the strength of a weld should depend on the relative importance of the processes.

Guerra, M.↗

Material Flow During Friction Stir Welds

The flow of metal during Friction Stir Welding is clarified using a faying surface tracer and a nib frozen in place during welding. It is shown that material is transported in two distinct streams or currents. One stream is a wiping of material from the advancing front side of the nib onto a plug of material that rotates and advances with the nib. The material undergoes a helical motion within the plug that both rotates and advances with the plug and descends in the wash of the threads on the nib and rises on the outer part of the plug. After one or more rotations, this material is sloughed off the plug in the wake of the tool primarily on the advancing side. The second stream of material is an entrainment of material from the retreating side of the nib that fills in between the sloughed off pieces from the advancing side. These two processes produce material with different mechanical properties and the strength of a weld should depend on the relative importance of the processes.

Guerra, M.↗

Spinning Rocket Simulator Turntable Design

Contained herein is the research and data acquired from the Turntable Design portion of the Spinning Rocket Simulator (SRS) project. The SRS Project studies and eliminates the effect of coning on thrust-propelled spacecraft. This design and construction of the turntable adds a structural support for the SRS model and two degrees of freedom. The two degrees of freedom, radial and circumferential, will help develop a simulated thrust force perpendicular to the plane of the spacecraft model while undergoing an unstable coning motion. The Turntable consists of a ten-foot linear track mounted to a sprocket and press-fit to a thrust bearing. A two-inch high column grounded by a Triangular Baseplate supports this bearing and houses the slip rings and pressurized, air-line swivel. The thrust bearing allows the entire system to rotate under the moment applied through the chain-driven sprocket producing a circumferential degree of freedom. The radial degree of freedom is given to the model through the helically threaded linear track. This track allows the Model Support and Counter Balance to simultaneously reposition according to the coning motion of the Model. Two design factors that hinder the linear track are bending and twist due to torsion. A Standard Aluminum "C" channel significantly reduces these two deflections. Safety considerations dictate the design of all the components involved in this project.

Miles, Robert W.↗

Automation of Data Traffic Control on DSM Architecture

The design of distributed shared memory (DSM) computers liberates users from the duty to distribute data across processors and allows for the incremental development of parallel programs using, for example, OpenMP or Java threads. DSM architecture greatly simplifies the development of parallel programs having good performance on a few processors. However, to achieve a good program scalability on DSM computers requires that the user understand data flow in the application and use various techniques to avoid data traffic congestions. In this paper we discuss a number of such techniques, including data blocking, data placement, data transposition and page size control and evaluate their efficiency on the NAS (NASA Advanced Supercomputing) Parallel Benchmarks. We also present a tool which automates the detection of constructs causing data congestions in Fortran array oriented codes and advises the user on code transformations for improving data traffic in the application.

Frumkin, Michael↗

ScienceOrganizer System and Interface Summary

ScienceOrganizer is a specialized knowledge management tool designed to enhance the information storage, organization, and access capabilities of distributed NASA science teams. Users access ScienceOrganizer through an intuitive Web-based interface that enables them to upload, download, and organize project information - including data, documents, images, and scientific records associated with laboratory and field experiments. Information in ScienceOrganizer is "threaded", or interlinked, to enable users to locate, track, and organize interrelated pieces of scientific data. Linkages capture important semantic relationships among information resources in the repository, and these assist users in navigating through the information related to their projects.

Keller, Richard M.↗

Motions and Initial Conditions in Star-Forming Dense Cores

Under this grant in the past year we have pursued spectral-line observations of star-forming regions over size scales from 0.01 pc to 0.5 pc. Our main goal has been to measure the systematic and turbulent motions of condensing and collapsing gas. In this area, our results include (1) in 67 starless dense cores, some 19 show clear evidence of spatially extended inward motions, with typical line-of-sight inward speed 0.05-0.09 km s(sup -1) and with typical plane-of-the-sky extent 0.1-0.3 pc, (2) In some 40 nearby regions with embedded groups and clusters, we see extended infall asymmetry in lines of CS and HCO(+) clearly in 4 regions and less clearly in 4 others, (3) Using finer resolution (15 arcsec or 0.01-0.02 pc) and lines tracing higher density, we see spatial concentration of infall asymmetry near the protostars in NGC 1333 IRS 4A and B, L483, and L1251B, and with still finer resolution (2 arcsec or 0.003 pc or 600 AU) we detect inverse P Cyg profiles, indicating absorption of continuum emission from the protostellar envelope by infalling gas in NGC 1333 IRS 4A and 4B. Further, at high resolution we identify regions of stellar mass and low turbulence ("kernels") which are good candidates to become the next generation of stars in embedded clusters. In addition we have completed a survey for the OH Zeeman effect in absorption against nearby H II regions, indicating that the large-scale magnetic field may be nearly critical if it typically threads a flattened structure. We have also developed a model of spatially extended infall motions based on dissipation of turbulence in a magnetized, selfgravitating layer. In the following we describe some of these results in more detail.

Myers, Philip C.↗

Multiple Core Galaxies

Nuclei of galaxies often show complicated density structures and perplexing kinematic signatures. In the past we have reported numerical experiments indicating a natural tendency for galaxies to show nuclei offset with respect to nearby isophotes and for the nucleus to have a radial velocity different from the galaxy's systemic velocity. Other experiments show normal mode oscillations in galaxies with large amplitudes. These oscillations do not damp appreciably over a Hubble time. The common thread running through all these is that galaxies often show evidence of ringing, bouncing, or sloshing around in unexpected ways, even though they have not been disturbed by any external event. Recent observational evidence shows yet another phenomenon indicating the dynamical complexity of central regions of galaxies: multiple cores (M31, Markarian 315 and 463 for example). These systems can hardly be static. We noted long-lived multiple core systems in galaxies in numerical experiments some years ago, and we have more recently followed up with a series of experiments on multiple core galaxies, starting with two cores. The relevant parameters are the energy in the orbiting clumps, their relative.masses, the (local) strength of the potential well representing the parent galaxy, and the number of cores. We have studied the dependence of the merger rates and the nature of the final merger product on these parameters. Individual cores survive much longer in stronger background potentials. Cores can survive for a substantial fraction of a Hubble time if they travel on reasonable orbits.

Miller, R.H.↗

Aerosol Optical Properties over the Oceans: Summary and Interpretation of Shadow-Band Radiometer Data from Six Cruises

The aerosol scattering component of the total radiance measured at the detectors of ocean color satellites is determined with atmospheric correction algorithms. These algorithms are based on aerosol optical thickness measurements made in two channels that lie in the near-infrared portion of the electromagnetic spectrum. The aerosol properties in the near-infrared region are used because there is no significant contribution to the satellite-measured radiance from the underlying ocean surface in that spectral region. In the visible wavelength bands, the spectrum of radiation scattered from the turbid atmosphere is convolved with the spectrum of radiation scattered from the surface layers of the ocean. The radiance contribution made by aerosols in the visible bands is determined from the near-infrared measurements through the use of aerosol models and radiation transfer codes. Selection of appropriate aerosol models from the near-infrared measurements is a fundamental challenge. There are several challenges with respect to the development, improvement, and evaluation of satellite ocean-color atmospheric correction algorithms. A common thread among these challenges is the lack of over-ocean aerosol data. Until recently, one of the most important limitations has been the lack of techniques and instruments to make aerosol measurements at sea. There has been steady progress in this area over the past five years, and there are several new and promising devices and techniques for data collection. The development of new instruments and the collection of more aerosol data from over the world's oceans have brought the realization that aerosol measurements that can be directly compared with aerosol measurements from ocean color satellite measurements are difficult to obtain. There are two problems that limit these types of comparisons: the cloudiness of the atmosphere over the world's oceans and the limitations of the techniques and instruments used to collect aerosol data from ships. To address the latter, we have developed a new type of shipboard sun photometer.

Miller, Mark A.↗

Cooperative Data Sharing: Simple Support for Clusters of SMP Nodes

Libraries like PVM and MPI send typed messages to allow for heterogeneous cluster computing. Lower-level libraries, such as GAM, provide more efficient access to communication by removing the need to copy messages between the interface and user space in some cases. still lower-level interfaces, such as UNET, get right down to the hardware level to provide maximum performance. However, these are all still interfaces for passing messages from one process to another, and have limited utility in a shared-memory environment, due primarily to the fact that message passing is just another term for copying. This drawback is made more pertinent by today's hybrid architectures (e.g. clusters of SMPs), where it is difficult to know beforehand whether two communicating processes will share memory. As a result, even portable language tools (like HPF compilers) must either map all interprocess communication, into message passing with the accompanying performance degradation in shared memory environments, or they must check each communication at run-time and implement the shared-memory case separately for efficiency. Cooperative Data Sharing (CDS) is a single user-level API which abstracts all communication between processes into the sharing and access coordination of memory regions, in a model which might be described as "distributed shared messages" or "large-grain distributed shared memory". As a result, the user programs to a simple latency-tolerant abstract communication specification which can be mapped efficiently to either a shared-memory or message-passing based run-time system, depending upon the available architecture. Unlike some distributed shared memory interfaces, the user still has complete control over the assignment of data to processors, the forwarding of data to its next likely destination, and the queuing of data until it is needed, so even the relatively high latency present in clusters can be accomodated. CDS does not require special use of an MMU, which can add overhead to some DSM systems, and does not require an SPMD programming model. unlike some message-passing interfaces, CDS allows the user to implement efficient demand-driven applications where processes must "fight" over data, and does not perform copying if processes share memory and do not attempt concurrent writes. CDS also supports heterogeneous computing, dynamic process creation, handlers, and a very simple thread-arbitration mechanism. Additional support for array subsections is currently being considered. The CDS1 API, which forms the kernel of CDS, is built primarily upon only 2 communication primitives, one process initiation primitive, and some data translation (and marshalling) routines, memory allocation routines, and priority control routines. The entire current collection of 28 routines provides enough functionality to implement most (or all) of MPI 1 and 2, which has a much larger interface consisting of hundreds of routines. still, the API is small enough to consider integrating into standard os interfaces for handling inter-process communication in a network-independent way. This approach would also help to solve many of the problems plaguing other higher-level standards such as MPI and PVM which must, in some cases, "play OS" to adequately address progress and process control issues. The CDS2 API, a higher level of interface roughly equivalent in functionality to MPI and to be built entirely upon CDS1, is still being designed. It is intended to add support for the equivalent of communicators, reduction and other collective operations, process topologies, additional support for process creation, and some automatic memory management. CDS2 will not exactly match MPI, because the copy-free semantics of communication from CDS1 will be supported. CDS2 application programs will be free to carefully also use CDS1. CDS1 has been implemented on networks of workstations running unmodified Unix-based operating systems, using UDP/IP and vendor-supplied high- performance locks. Although its inter-node performance is currently unimpressive due to rudimentary implementation technique, it even now outperforms highly-optimized MPI implementation on intra-node communication due to its support for non-copy communication. The similarity of the CDS1 architecture to that of other projects such as UNET and TRAP suggests that the inter-node performance can be increased significantly to surpass MPI or PVM, and it may be possible to migrate some of its functionality to communication controllers.

DiNucci, David C.↗

Flow Trajectories in Friction Stir Welding

In the Friction Stir Welding (FSW) process, a rotating, shouldered tool with a threaded pin is inserted under pressure along the seam of two pieces of metal which are tightly clamped together, and secured against a rigid anvil underneath. The rotating pin travels along the seam and through a combination of pressure and friction heating produces a zone of plastic deformation around the pin within the workpiece on either side of the seam. As the pin is moved in the direction of welding, the plasticised material moves around the tool and bonds together behind it. The elements of the material flow behavior are a combination of three elements. There is a rotational transport of material being carried around the tool, extrusion of material being forced around the pin on both sides into the cavity created behind it, and a lifting and dropping of material as it is stirred and mixed by the rotating action of the pin. It was assumed that rotational motion of the plastic zone is the primary mechanism for transport of material around the welding tool. A kinematic mathematical model was used to compute trajectories of material movement for various distributions of rotational slip within the plastic zone. These trajectories were then compared with the results of an experiment that produced radiographs of markers embedded in a workpiece that was welded with the FSW process. It was assumed that the copper wire markers retained their original length as the aluminum material flowed around them. The kinematic model included a constraint so that the displacements were such that the total length of the wire markers did not increase after deformation. There was good agreement between the calculated trajectories for the case of localized slip at the outer surface of the plastic deformation zone and the radiographs of the copper wire markers. The trajectories differed markedly from the radiograph traces when a distributed slip zone was assumed. It was concluded that the flow field could be described generally by a rotating plug model with slip occurring on the surface at the outer boundary of the plastic zone.

Bernstein, Edward L.↗

Parallelization of the Flow Field Dependent Variation Scheme for Solving the Triple Shock/Boundary Layer Interaction Problem

A parallelized version of the Flowfield Dependent Variation (FDV) Method is developed to analyze a problem of current research interest, the flowfield resulting from a triple shock/boundary layer interaction. Such flowfields are often encountered in the inlets of high speed air-breathing vehicles including the NASA Hyper-X research vehicle. In order to resolve the complex shock structure and to provide adequate resolution for boundary layer computations of the convective heat transfer from surfaces inside the inlet, models containing over 500,000 nodes are needed. Efficient parallelization of the computation is essential to achieving results in a timely manner. Results from a parallelization scheme, based upon multi-threading, as implemented on multiple processor supercomputers and workstations is presented.

Schunk, Richard Gregory↗

Automatic Multilevel Parallelization Using OpenMP

In this paper we describe the extension of the CAPO (CAPtools (Computer Aided Parallelization Toolkit) OpenMP) parallelization support tool to support multilevel parallelism based on OpenMP directives. CAPO generates OpenMP directives with extensions supported by the NanosCompiler to allow for directive nesting and definition of thread groups. We report some results for several benchmark codes and one full application that have been parallelized using our system.

Jin, Hao-Qiang↗

Effects of Ordering Strategies and Programming Paradigms on Sparse Matrix Computations

The Conjugate Gradient (CG) algorithm is perhaps the best-known iterative technique to solve sparse linear systems that are symmetric and positive definite. For systems that are ill-conditioned, it is often necessary to use a preconditioning technique. In this paper, we investigate the effects of various ordering and partitioning strategies on the performance of parallel CG and ILU(O) preconditioned CG (PCG) using different programming paradigms and architectures. Results show that for this class of applications: ordering significantly improves overall performance on both distributed and distributed shared-memory systems, that cache reuse may be more important than reducing communication, that it is possible to achieve message-passing performance using shared-memory constructs through careful data ordering and distribution, and that a hybrid MPI+OpenMP paradigm increases programming complexity with little performance gains. A implementation of CG on the Cray MTA does not require special ordering or partitioning to obtain high efficiency and scalability, giving it a distinct advantage for adaptive applications; however, it shows limited scalability for PCG due to a lack of thread level parallelism.

Oliker, Leonid↗

Secure Future

Thread Technology, Inc., developed the ZipNut, a fastening device that is pushed rather than turned. It was originally developed for space shuttle and space station applications. The ZipNut can be used for a variety of applications, including firefighting, aerospace, gas fittings, and manufacturing.

Source record↗