Search NASA⌕ Search

SEARCH · Search NASA

Results for “program architectures”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 649 records · Page 36

The Kokkos OpenMPTarget Backend: Implementation and Lessons Learned

As the supercomputing landscape diversifies, solutions such as Kokkos to write vendor agnostic applications and libraries have risen in popularity. Kokkos provides a programming model designed for performance portability, which allows developers to write a single source implementation that can run efficiently on various architectures. At its heart, Kokkos maps parallel algorithms to architecture and vendor specific backends written in lower level programming models such as CUDA and HIP. Another approach to writing vendor agnostic parallel code is using OpenMP’s directives based approach, which lets developers annotate code to express parallelism. It is implemented at the compiler level and is supported by all major high performance computing vendors, as well as the primary Open Source toolchains GNU and LLVM. Since its inception, Kokkos has used OpenMP to parallelize on CPU architectures. In this paper, we explore leveraging OpenMP for a GPU backend and discuss the challenges we encountered when mapping the Kokkos APIs and semantics to OpenMP target constructs. As an exemplar workload we chose a simple conjugate gradient solver for sparse matrices. We find that performance on NVIDIA and AMD GPUs varies widely based on details of the implementation strategy and the chosen compiler. Furthermore, the performance of the OpenMP implementations decreases with increasing complexity of the investigated algorithms.

Gayatri, Rahulkumar↗

GPU COMPUTING FOR PARTICLE TRACKING

This is a feasibility study of using a modern Graphics Processing Unit (GPU) to parallelize the accelerator particle tracking code. To demonstrate the massive parallelization features provided by GPU computing, a simplified TracyGPU program is developed for dynamic aperture calculation. Performances, issues, and challenges from introducing GPU are also discussed. General purpose Computation on Graphics Processing Units (GPGPU) bring massive parallel computing capabilities to numerical calculation. However, the unique architecture of GPU requires a comprehensive understanding of the hardware and programming model to be able to well optimize existing applications. In the field of accelerator physics, the dynamic aperture calculation of a storage ring, which is often the most time consuming part of the accelerator modeling and simulation, can benefit from GPU due to its embarrassingly parallel feature, which fits well with the GPU programming model. In this paper, we use the Tesla C2050 GPU which consists of 14 multi-processois (MP) with 32 cores on each MP, therefore a total of 448 cores, to host thousands ot threads dynamically. Thread is a logical execution unit of the program on GPU. In the GPU programming model, threads are grouped into a collection of blocks Within each block, multiple threads share the same code, and up to 48 KB of shared memory. Multiple thread blocks form a grid, which is executed as a GPU kernel. A simplified code that is a subset of Tracy++ [2] is developed to demonstrate the possibility of using GPU to speed up the dynamic aperture calculation by having each thread track a particle.

42 ENGINEERING↗

Reverse time migration: A seismic processing application on the connection machine

The implementation of a reverse time migration algorithm on the Connection Machine, a massively parallel computer is described. Essential architectural features of this machine as well as programming concepts are presented. The data structures and parallel operations for the implementation of the reverse time migration algorithm are described. The algorithm matches the Connection Machine architecture closely and executes almost at the peak performance of this machine.

Fiebrich, Rolf-Dieter↗

Mission Options Scoping Tool for Mars Orbiters: Mass Cost Calculator (MC2)

Prior to developing the details of an advanced mission study, the mission architecture trade space is typically explored to assess the scope of feasible options. This paper describes the main features of an Excel-based tool, called the Mass-Cost-Calculator (MC2 ), which is used to perform rapid, high-level mass and cost options analyses of Mars orbiter missions. MC2 consists of a combination of databases, analytical solutions, and parametric relationships to enable quick evaluation of new mission concepts and comparison of multiple architecture options. The tool's outputs provide program management and planning teams with answers to "what if" queries, as well as an understanding of the driving mission elements, during the pre-project planning phase. These outputs have been validated against the outputs generated by the Advanced Projects Design Team (Team X) at NASA's Jet Propulsion Laboratory (JPL). The architecture of the tool allows for future expansion to other orbiters beyond Mars, and to non-orbiter missions, such as those involving fly-by spacecraft, probes, landers, rovers, or other mission elements.

orbiters↗

Architecture and properties of TCR fuel form

The fuel form developed for the Transformational Challenge Reactor demonstration program leverages recent advances in manufacturing, materials, and computational sciences, delivering a new architecture for production of high-performance microencapsulated nuclear fuels. The fuel consists of conventionally manufactured uranium nitride tristructural isotropic fuel particles embedded inside a 3D-printed silicon carbide matrix. Finally, this paper describes the overall architecture and manufacturing process for this fuel form, its properties and behavior, and the ongoing development activities.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Data storage technology: Hardware and software, Appendix B

This project involves the development of more economical ways of integrating and interfacing new storage devices and data processing programs into a computer system. It involves developing interface standards and a software/hardware architecture which will make it possible to develop machine independent devices and programs. These will interface with the machine dependent operating systems of particular computers. The development project will not be to develop the software which would ordinarily be the responsibility of the manufacturer to supply, but to develop the standards with which that software is expected to confirm in providing an interface with the user or storage system.

Sable, J. D.↗

The Design and Evaluation of "CAPTools"--A Computer Aided Parallelization Toolkit

Writing applications for high performance computers is a challenging task. Although writing code by hand still offers the best performance, it is extremely costly and often not very portable. The Computer Aided Parallelization Tools (CAPTools) are a toolkit designed to help automate the mapping of sequential FORTRAN scientific applications onto multiprocessors. CAPTools consists of the following major components: an inter-procedural dependence analysis module that incorporates user knowledge; a 'self-propagating' data partitioning module driven via user guidance; an execution control mask generation and optimization module for the user to fine tune parallel processing of individual partitions; a program transformation/restructuring facility for source code clean up and optimization; a set of browsers through which the user interacts with CAPTools at each stage of the parallelization process; and a code generator supporting multiple programming paradigms on various multiprocessors. Besides describing the rationale behind the architecture of CAPTools, the parallelization process is illustrated via case studies involving structured and unstructured meshes. The programming process and the performance of the generated parallel programs are compared against other programming alternatives based on the NAS Parallel Benchmarks, ARC3D and other scientific applications. Based on these results, a discussion on the feasibility of constructing architectural independent parallel applications is presented.

Yan, Jerry↗

Parallel Programming Models and Systems for High Performance Computing. In: Emerging Research in Cloud Distributed Computing Systems

A parallel programming model is an abstraction of a parallel system that allows expression of both algorithms and shared data structures. To accommodate the diversity in parallel system architectures and user requirements, there are a variety of programming models including the models providing a shared memory view or a distributed memory view of the system. The programming models are implemented as libraries, language extensions, or compiler directives. This chapter provides a discussion on programming models and its implementations aimed at application developers, system software researchers, and hardware architects. The first part provides an overview of the programming models. The second part is an in-depth discussion on high-performance networking interface to implement the programming model. The last part of the chapter discusses implementation of a programming model with a case study. Each part of the chapter concludes with a discussion on current research trends and its impact on future architectures.

Venkata, Manjunath Gorentia↗

Integrated Framework to Enable Design Space Exploration of In-Space Transportation Architectures

Even with the recent shift in focus of NASA’s human spaceflight program towards the Moon, the long-term goal continues to be crewed missions to Mars. Regardless of the accepted architecture, in-space transportation systems are a critical portion to achieving this goal. An emphasis is placed on presenting decision makers with many options early on to make an informed down-selection. Although current processes are able to address any aspects of the challenging underlying multi-disciplinary analysis and optimization problem of these advanced concepts, they are unable to comprehensively perform design space exploration inexpensively,potentially leaving attractive alternatives on the table. A design framework is proposed that is capable of enabling the integrated mission analysis, vehicle sizing and synthesis problem for in-space transportation architectures. This framework is enabled through the use of the vehicle synthesis tool, Dynamic Rocket Equation Tool, various subsystem sizing models, and low-thrust trajectory optimization codes. As a case study to demonstrate the framework, two design points of a hybrid propulsion stage concept, studied by Georgia Tech’s 2018 Revolutionary Aerospace Systems Concepts-Academic Linkage team, is explored.

Akshay Prasad↗

Integrated Framework to Enable Design Space Exploration of In-Space Transportation Architectures

Even with the recent shift in focus of NASA’s human spaceflight program towards the Moon, the long-term goal continues to be crewed missions to Mars. Regardless of the accepted architecture, in-space transportation systems are a critical portion to achieving this goal. An emphasis is placed on presenting decision makers with many options early on to make an informed down-selection. Although current processes are able to address any aspects of the challenging underlying multi-disciplinary analysis and optimization problem of these advanced concepts, they are unable to comprehensively perform design space exploration inexpensively, potentially leaving attractive alternatives on the table. A design framework is proposed that is capable of enabling the integrated mission analysis, vehicle sizing and synthesis problem for in-space transportation architectures. This framework is enabled through the use of the vehicle synthesis tool, Dynamic Rocket Equation Tool, various subsystem sizing models, and low-thrust trajectory optimization codes. As a case study to demonstrate the framework, two design points of a hybrid propulsion stage concept, studied by Georgia Tech’s 2018 Revolutionary Aerospace Systems Concepts-Academic Linkage team, is explored.

Akshay Prasad↗

Reinforcement-based Program Induction in a Neural Virtual Machine

We present a neural virtual machine that can be trained to perform algorithmic tasks. Rather than combining a neural controller with non-neural memory storage as has been done in the past, this architecture is purely neural and emulates tape-based memory via fast associative weights (onestep learning). Here we formally define the architecture, and then extend the system to learn programs using recurrent policy gradient reinforcement learning based on examples of program inputs labeled with corresponding output targets, which are compared against actual output to generate a sparse reward signal. We describe the policy gradient training procedure used, and report its empirical performance on a number of smallscale list processing tasks, such as finding the maximum list element, filtering out certain elements, and reversing the order of the elements. These results show that program induction via reinforcement learning is possible using sparse rewards and solely neural computations.

Katz, Garrett E.↗

Verifying quantitative reliability for programs that execute on unreliable hardware

Emerging high-performance architectures are expected to contain unreliable components that may exhibit soft errors, which silently corrupt the results of computations. Full detection and masking of soft errors is challenging, expensive, and, for some applications, unnecessary. For example, approximate computing applications (such as multimedia processing, machine learning, and big data analytics) can often naturally tolerate soft errors. We introduce Rely, a programming language that enables developers to reason about the quantitative reliability of an application—namely, the probability that it produces the correct result when executed on unreliable hardware. Rely allows developers to specify the reliability requirements for each value that a function produces. We perform a static quantitative reliability analysis that verifies quantitative requirements on the reliability of an application, enabling a developer to perform sound and verified reliability engineering. The analysis takes a Rely program with a reliability specification and a hardware specification that characterizes the reliability of the underlying hardware components and verifies that the program satisfies its reliability specification when executed on the underlying unreliable hardware platform. We demonstrate the application of quantitative reliability analysis on six computations implemented in Rely.

97 MATHEMATICS AND COMPUTING↗

A visual programming environment for the Navier-Stokes computer

The Navier-Stokes computer is a high-performance, reconfigurable, pipelined machine designed to solve large computational fluid dynamics problems. Due to the complexity of the architecture, development of effective, high-level language compilers for the system appears to be a very difficult task. Consequently, a visual programming methodology has been developed which allows users to program the system at an architectural level by constructing diagrams of the pipeline configuration. These schematic program representations can then be checked for validity and automatically translated into machine code. The visual environment is illustrated by using a prototype graphical editor to program an example problem.

Tomboulian, Sherryl↗

Northwest Open Automated Demand Response Technology Demonstration Project

The Lawrence Berkeley National Laboratory (LBNL) Demand Response Research Center (DRRC) demonstrated and evaluated open automated demand response (OpenADR) communication infrastructure to reduce winter morning and summer afternoon peak electricity demand in commercial buildings the Seattle area. LBNL performed this demonstration for the Bonneville Power Administration (BPA) in the Seattle City Light (SCL) service territory at five sites: Seattle Municipal Tower, Seattle University, McKinstry, and two Target stores. This report describes the process and results of the demonstration. OpenADR is an information exchange model that uses a client-server architecture to automate demand-response (DR) programs. These field tests evaluated the feasibility of deploying fully automated DR during both winter and summer peak periods. DR savings were evaluated for several building systems and control strategies. This project studied DR during hot summer afternoons and cold winter mornings, both periods when electricity demand is typically high. This is the DRRC project team's first experience using automation for year-round DR resources and evaluating the flexibility of commercial buildings end-use loads to participate in DR in dual-peaking climates. The lessons learned contribute to understanding end-use loads that are suitable for dispatch at different times of the year. The project was funded by BPA and SCL. BPA is a U.S. Department of Energy agency headquartered in Portland, Oregon and serving the Pacific Northwest. BPA operates an electricity transmission system and markets wholesale electrical power at cost from federal dams, one non-federal nuclear plant, and other non-federal hydroelectric and wind energy generation facilities. Created by the citizens of Seattle in 1902, SCL is the second-largest municipal utility in America. SCL purchases approximately 40% of its electricity and the majority of its transmission from BPA through a preference contract. SCL also provides ancillary services within its own balancing authority. The relationship between BPA and SCL creates a unique opportunity to create DR programs that address both BPA's and SCL's markets simultaneously. Although simultaneously addressing both market could significantly increase the value of DR programs for BPA, SCL, and the end user, establishing program parameters that maximize this value is challenging because of complex contractual arrangements and the absence of a central Independent System Operator or Regional Transmission Organization in the northwest.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Advanced information processing system: Authentication protocols for network communication

In safety critical I/O and intercomputer communication networks, reliable message transmission is an important concern. Difficulties of communication and fault identification in networks arise primarily because the sender of a transmission cannot be identified with certainty, an intermediate node can corrupt a message without certainty of detection, and a babbling node cannot be identified and silenced without lengthy diagnosis and reconfiguration . Authentication protocols use digital signature techniques to verify the authenticity of messages with high probability. Such protocols appear to provide an efficient solution to many of these problems. The objective of this program is to develop, demonstrate, and evaluate intercomputer communication architectures which employ authentication. As a context for the evaluation, the authentication protocol-based communication concept was demonstrated under this program by hosting a real-time flight critical guidance, navigation and control algorithm on a distributed, heterogeneous, mixed redundancy system of workstations and embedded fault-tolerant computers.

Harper, Richard E.↗

Architecting a Human Spaceflight Program

Presentation reviews: (1) What is Architecting, (2) Constellation "Architecture", (3) Human Exploration Architecture (HEFT) Architecting Effort, (4) Forward work in Human Spaceflight Architecting

Human Exploration Architecture (HEFT)↗

MIDEX Advanced Modular and Distributed Spacecraft Avionics Architecture

MIDEX (Medium Class Explorer) is the newest line in NASA's Explorer spacecraft development program. As part of the MIDEX charter, the MIDEX spacecraft development team has developed a new modular, distributed, and scaleable spacecraft architecture that pioneers new spaceflight technologies and implementation approaches, all designed to reduce overall spacecraft cost while increasing overall functional capability. This resultant "plug and play" system dramatically decreases the complexity and duration of spacecraft integration and test, providing a basic framework that supports spacecraft modularity and scalability for missions of varying size and complexity. Together, these subsystems form a modular, flexible avionics suite that can be modified and expanded to support low-end and very high-end mission requirements with a minimum of redesign, as well as allowing a smooth, continuous infusion of new technologies as they are developed without redesigning the system. This overall approach has the net benefit of allowing a greater portion of the overall mission budget to be allocated to mission science instead of a spacecraft bus. The MIDEX scaleable architecture is currently being manufactured and tested for use on the Microwave Anisotropy Probe (MAP), an inhouse program at GSFC.

Ruffa, John A.↗

Probing for exoplanets hiding in dusty debris disks: Disk imaging, characterization, and exploration with HST/STIS multi-roll coronagraphy

Spatially resolved scattered-light images of circumstellar debris in exoplanetary systems constrain the physical properties and orbits of the dust particles in these systems. They also inform on co-orbiting (but unseen) planets, the systemic architectures, and forces perturbing the starlight-scattering circumstellar material. Using Hubble Space Telescope (HST)/Space Telescope Imaging Spectrograph (STIS) broadband optical coronagraphy, we have completed the observational phase of a program to study the spatial distribution of dust in a sample of 10 circumstellar debris systems and 1 'mature' protoplanetrary disk, all with HST pedigree, using point-spread-function-subtracted multi-roll coronagraphy. These observations probe stellocentric distances ≥5 AU for the nearest systems, and simultaneously resolve disk substructures well beyond corresponding to the giant planet and Kuiper Belt regions within our own solar system. They also disclose diffuse very low-surface-brightness dust at larger stellocentric distances. Herein we present new results inclusive of fainter disks such as HD 92945 (F {sub disk}/F {sub star} = 5 × 10{sup –5}), confirming, and better revealing, the existence of a narrow inner debris ring within a larger diffuse dust disk. Other disks with ring-like substructures and significant asymmetries and complex morphologies include HD 181327, for which we posit a spray of ejecta from a recent massive collision in an exo-Kuiper Belt; HD 61005, suggested to be interacting with the local interstellar medium; and HD 15115 and HD 32297, also discussed in the context of putative environmental interactions. These disks and HD 15745 suggest that debris system evolution cannot be treated in isolation. For AU Mic's edge-on disk, we find out-of-plane surface brightness asymmetries at ≥5 AU that may implicate the existence of one or more planetary perturbers. Time-resolved images of the MP Mus protoplanetary disk provide spatially resolved temporal variability in the disk illumination. These and other new images from our HST/STIS GO/12228 program enable direct inter-comparison of the architectures of these exoplanetary debris systems in the context of our own solar system.

79 ASTRONOMY AND ASTROPHYSICS↗