SEARCH · Search NASA

Results for “static graph”

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.

SourceResearch connectionFrom yearThrough year

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

At least 19 records

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Towards Enhancing Coding Productivity for GPU Programming Using Static Graphs

The main contribution of this work is to increase the coding productivity of GPU programming by using the concept of Static Graphs. GPU capabilities have been increasing significantly in terms of performance and memory capacity. However, there are still some problems in terms of scalability and limitations to the amount of work that a GPU can perform at a time. To minimize the overhead associated with the launch of GPU kernels, as well as to maximize the use of GPU capacity, we have combined the new CUDA Graph API with the CUDA programming model (including CUDA math libraries) and the OpenACC programming model. We use as test cases two different, well-known and widely used problems in HPC and AI: the Conjugate Gradient method and the Particle Swarm Optimization. In the first test case (Conjugate Gradient) we focus on the integration of Static Graphs with CUDA. In this case, we are able to significantly outperform the NVIDIA reference code, reaching an acceleration of up to 11x thanks to a better implementation, which can benefit from the new CUDA Graph capabilities. In the second test case (Particle Swarm Optimization), we complement the OpenACC functionality with the use of CUDA Graph, achieving again accelerations of up to one order of magnitude, with average speedups ranging from 2x to 4x, and performance very close to a reference and optimized CUDA code. Our main target is to achieve a higher coding productivity model for GPU programming by using Static Graphs, which provides, in a very transparent way, a better exploitation of the GPU capacity. The combination of using Static Graphs with two of the current most important GPU programming models (CUDA and OpenACC) is able to reduce considerably the execution time w.r.t. the use of CUDA and OpenACC only, achieving accelerations of up to more than one order of magnitude. Finally, we propose an interface to incorporate the concept of Static Graphs into the OpenACC Specifications.

58 GEOSCIENCES↗

DOE OSTI2021

Static Graphs for Coding Productivity in OpenACC

The main contribution of this work is to increase the coding productivity for GPU programming by using the concept of Static Graphs. To do so, we have combined the new CUDA Graph API with the OpenACC programming model. We use as test cases a well-known and widely used problems in HPC and AI: the Particle Swarm Optimization. We complement the OpenACC functionality with the use of CUDA Graph, achieving accelerations of more than one order of magnitude, and a performance very close to a reference and optimized CUDA code. Finally, we propose a new specification to incorporate the concept of Static Graphs into the OpenACC specification.

Toledo, Leonel↗

DOE OSTI2023

Exploring temporal community evolution: algorithmic approaches and parallel optimization for dynamic community detection

Abstract Dynamic (temporal) graphs are a convenient mathematical abstraction for many practical complex systems including social contacts, business transactions, and computer communications. Community discovery is an extensively used graph analysis kernel with rich literature for static graphs. However, community discovery in a dynamic setting is challenging for two specific reasons. Firstly, the notion of temporal community lacks a widely accepted formalization, and only limited work exists on understanding how communities emerge over time. Secondly, the added temporal dimension along with the sheer size of modern graph data necessitates new scalable algorithms. In this paper, we investigate how communities evolve over time based on several graph metrics under a temporal formalization. We compare six different algorithmic approaches for dynamic community detection for their quality and runtime. We identify that a vertex-centric (local) optimization method works as efficiently as the classical modularity-based methods. To its advantage, such local computation allows for the efficient design of parallel algorithms without incurring a significant parallel overhead. Based on this insight, we design a shared-memory parallel algorithm DyComPar , which demonstrates between 4 and 18 fold speed-up on a multi-core machine with 20 threads, for several real-world and synthetic graphs from different domains.

97 MATHEMATICS AND COMPUTING↗

NASA NTRSDate not supplied

Towards Sheaf Theoretic Analyses for Delay Tolerant Networking

The goal of Delay Tolerant Networking (DTN) is to take a collection of heterogeneous, disparate connections between satellites, space assets, ground stations, and ground infrastructure and bring it together into a cohesive, functioning overlay network. Depending on the systems being considered, one can find links with a one-way light time exceeding minutes (and hours),periodic links which can sometimes be predicted by orbital mechanics, and restrictions based on the variety of capabilities built into these systems. These characteristics preclude traditional network models and routing techniques and have classically led to either rigid routing tables or purely probabilistic models. As the deeper underlying structures remain unknown, development of more DTN-optimized algorithms has lacked the necessary foundation. In a continuation of previous work, the goal of this paper is to identify and study these fundamental structures that exist in delay tolerant networks (DTN), with a focus on space networks. The current routing methodology has been to use contact graph routing (CGR) algorithms. CGR models a series of known contacts as a static graph. For CGR to work, this graph must be globally consistent and must have an accurate picture of the network. Because this is a globally controlled structure, there is little room for flexibility in the event of changes to the network which would naturally occur as the network grows. As a response to the desire for flexibility as the network changes, we introduced the mathematical structure known as sheaves to DTNs last year. The tag-line for sheaves is that they are a mathematically precise way of gluing local data together into unique global data. Thus, sheaves lend extra power to traditional models(and routing algorithms) by taking additional information and merging it, in as consistent a manner as possible, with the representation itself. The clearest example of how Earth-bound networks exhibit behavior that is “sheafy” is link state routers, which build a local-to-global picture of their network by gluing local information together into a global network, exactly as a sheaf would do. For routing within delay tolerant networks to truly exploit this structure, a deeper structure than a graph is required. In this paper, we develop sheaves that can work over directed graphs such as temporal flow networks, we construct a sheaf representation for Dijkstra’s algorithm, and we outline a construction for routing sheaves capable of modeling multicast scenarios. Finally, there is a section of future work suggesting follow-on research.

Robert Short↗

DOE OSTI2024

Anticipating Technical Expertise and Capability Evolution in Research Communities Using Dynamic Graph Transformers

The ability to anticipate global technical expertise and capability evolution trends is essential for national and global security, especially in safety-critical domains such as nuclear nonproliferation (NN) and rapidly emerging fields like artificial intelligence (AI). Here, in this work, we extend traditional statistical relational learning approaches (e.g., link prediction in collaboration networks) and formulate a problem of anticipating technical expertise and capability evolution using dynamic heterogeneous graph representations. We develop novel capabilities to forecast collaboration patterns, authorship behavior, and technical capability evolution at different granularities (e.g., scientist and institution levels) in two distinct research fields. We implement a dynamic graph transformer (DGT) neural architecture, which pushes the state-of-the-art graph neural network models by: 1) forecasting heterogeneous (rather than homogeneous) nodes and edges; and 2) relying on both discrete- and continuous-time inputs. We demonstrate that our DGT models predict collaboration, partnership, and expertise patterns with 0.26, 0.73, and 0.53 mean reciprocal rank values for AI and 0.48, 0.93, and 0.22 for NN domains. DGT model performance exceeds the best-performing static graph baseline models by 30%–80% across AI and NN domains. Our findings demonstrate that DGT models boost inductive task performance when previously unseen nodes appear in the test data for the domains with emerging collaboration patterns (e.g., AI). Specifically, models accurately predict which established scientists will collaborate with early career scientists and vice versa in the AI domain.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2005

A Comparison of Geographic Information Systems, Complex Networks, and Other Models for Analyzing Transportation Network Topologies

This report reviews six classes of models that are used for studying transportation network topologies. The report is motivated by two main questions. First, what can the "new science" of complex networks (scale-free, small-world networks) contribute to our understanding of transport network structure, compared to more traditional methods? Second, how can geographic information systems (GIS) contribute to studying transport networks? The report defines terms that can be used to classify different kinds of models by their function, composition, mechanism, spatial and temporal dimensions, certainty, linearity, and resolution. Six broad classes of models for analyzing transport network topologies are then explored: GIS; static graph theory; complex networks; mathematical programming; simulation; and agent-based modeling. Each class of models is defined and classified according to the attributes introduced earlier. The paper identifies some typical types of research questions about network structure that have been addressed by each class of model in the literature.

Alexandrov, Natalia↗

NASA NTRS1993

Performance analysis of a large-grain dataflow scheduling paradigm

A paradigm for scheduling computations on a network of multiprocessors using large-grain data flow scheduling at run time is described and analyzed. The computations to be scheduled must follow a static flow graph, while the schedule itself will be dynamic (i.e., determined at run time). Many applications characterized by static flow exist, and they include real-time control and digital signal processing. With the advent of computer-aided software engineering (CASE) tools for capturing software designs in dataflow-like structures, macro-dataflow scheduling becomes increasingly attractive, if not necessary. For parallel implementations, using the macro-dataflow method allows the scheduling to be insulated from the application designer and enables the maximum utilization of available resources. Further, by allowing multitasking, processor utilizations can approach 100 percent while they maintain maximum speedup. Extensive simulation studies are performed on 4-, 8-, and 16-processor architectures that reflect the effects of communication delays, scheduling delays, algorithm class, and multitasking on performance and speedup gains.

Young, Steven D.↗

NASA NTRS1997

Sequential Test Strategies for Multiple Fault Isolation

In this paper, we consider the problem of constructing near optimal test sequencing algorithms for diagnosing multiple faults in redundant (fault-tolerant) systems. The computational complexity of solving the optimal multiple-fault isolation problem is super-exponential, that is, it is much more difficult than the single-fault isolation problem, which, by itself, is NP-hard. By employing concepts from information theory and Lagrangian relaxation, we present several static and dynamic (on-line or interactive) test sequencing algorithms for the multiple fault isolation problem that provide a trade-off between the degree of suboptimality and computational complexity. Furthermore, we present novel diagnostic strategies that generate a static diagnostic directed graph (digraph), instead of a static diagnostic tree, for multiple fault diagnosis. Using this approach, the storage complexity of the overall diagnostic strategy reduces substantially. Computational results based on real-world systems indicate that the size of a static multiple fault strategy is strictly related to the structure of the system, and that the use of an on-line multiple fault strategy can diagnose faults in systems with as many as 10,000 failure sources.

Shakeri, M.↗

DOE OSTI2025

Model-based Hierarchical Reinforcement Learning for Improved Physical Security Design: A Prototype

Prior work in FY24 developed an adversarial AI agent aid in path analysis of physical protection systems. This agent, trained using a model-based reinforcement learning algorithm, was able to successfully learn the most vulnerable path in facilities. It was able to extend the current state of practice for physical protection design by exhibiting dynamic behavior based on current environmental conditions. Whereas PathTrace largely performs a static, graph-based analysis, the AI agent was able to make decisions based on relative position in the facility, current conditions (was the adversarial agnet discovered?), and proximity to secondary targets. The agent demonstrated some novel capabilities, but had limitations that need to be resolved before it can be used for production purposes. For example, the adversarial agent generalizes poorly and takes a relatively long time to train. Nonetheless, there is still considerable promise for developing the adversarial agent further in order to explore even richer, more dynamic behaviors (e.g., adversary motivations, environmental debris, and more). This work considers a complementary idea; development of a planning agent. The planning agent is envisioned as an auto-complete-like tool that can help accelerate security system design by human experts. The agent would respect existing barriers and sensors placed by a human expert while offering cost-effective suggestions (i.e., implicitly balancing effectiveness with cost) to improve the design. The goal is for this agent to be part of an expert’s toolbox, not to totally upend the current state-of-practice, or to displace human experts. The ultimate goal would be concurrent training of both the adversarial and planning agent together, to learn entirely through self-play. This would represent an entirely new way of performing system deign. We selected a hierarchical, model-based reinforcement learning algorithm to serve as the planning agent. This is an extension of concepts used in the prior FY24 adversarial agent work. There, we had a single agent acting an environment. Here, we have two different sub-agents (policies), working together, to form a complete agent. There is a manager policy, which can select abstract goals on slower time scales, and a worker, which performs primitive actions to reach goals selected by the manager. It is worth noting that this class of algorithm is challenging to work with. From our understanding, our work is one of the first successful uses of model-based reinforcement learning (MBRL) in nuclear energy1 , and likely the first hierarchical model-based reinforcement learning application in nuclear energy. Further, this work is one of the first known attempts to apply AI to perform a design tasks in nuclear energy. Consequently, there were significant implementation challenges and the bulk of the work was focused on successful implementation and algorithm design. The results presented here are very low technology readiness level as a consequence of the lack of related literature, but still represent a significant step forward in the pursuit of applied AI for design.

42 ENGINEERING↗

NASA NTRS2003

Scale-free Graphs for General Aviation Flight Schedules

In the late 1990s a number of researchers noticed that networks in biology, sociology, and telecommunications exhibited similar characteristics unlike standard random networks. In particular, they found that the cummulative degree distributions of these graphs followed a power law rather than a binomial distribution and that their clustering coefficients tended to a nonzero constant as the number of nodes, n, became large rather than O(1/n). Moreover, these networks shared an important property with traditional random graphs as n becomes large the average shortest path length scales with log n. This latter property has been coined the small-world property. When taken together these three properties small-world, power law, and constant clustering coefficient describe what are now most commonly referred to as scale-free networks. Since 1997 at least six books and over 400 articles have been written about scale-free networks. In this manuscript an overview of the salient characteristics of scale-free networks. Computational experience will be provided for two mechanisms that grow (dynamic) scale-free graphs. Additional computational experience will be given for constructing (static) scale-free graphs via a tabu search optimization approach. Finally, a discussion of potential applications to general aviation networks is given.

Alexandov, Natalia M.↗

NASA NTRS1993

A survey of program slicing for software engineering

This research concerns program slicing which is used as a tool for program maintainence of software systems. Program slicing decreases the level of effort required to understand and maintain complex software systems. It was first designed as a debugging aid, but it has since been generalized into various tools and extended to include program comprehension, module cohesion estimation, requirements verification, dead code elimination, and maintainence of several software systems, including reverse engineering, parallelization, portability, and reuse component generation. This paper seeks to address and define terminology, theoretical concepts, program representation, different program graphs, developments in static slicing, dynamic slicing, and semantics and mathematical models. Applications for conventional slicing are presented, along with a prognosis of future work in this field.

Beck, Jon↗

NASA NTRS2016

Transmission Scheduling and Routing Algorithms for Delay Tolerant Networks

The challenges of data processing, transmission scheduling and routing within a space network present a multi-criteria optimization problem. Long delays, intermittent connectivity, asymmetric data rates and potentially high error rates make traditional networking approaches unsuitable. The delay tolerant networking architecture and protocols attempt to mitigate many of these issues, yet transmission scheduling is largely manually configured and routes are determined by a static contact routing graph. A high level of variability exists among the requirements and environmental characteristics of different missions, some of which may allow for the use of more opportunistic routing methods. In all cases, resource allocation and constraints must be balanced with the optimization of data throughput and quality of service. Much work has been done researching routing techniques for terrestrial-based challenged networks in an attempt to optimize contact opportunities and resource usage. This paper examines several popular methods to determine their potential applicability to space networks.

Space Networking↗

NASA NTRS1994

Modeling heterogeneous processor scheduling for real time systems

A new model is presented to describe dataflow algorithms implemented in a multiprocessing system. Called the resource/data flow graph (RDFG), the model explicitly represents cyclo-static processor schedules as circuits of processor arcs which reflect the order that processors execute graph nodes. The model also allows the guarantee of meeting hard real-time deadlines. When unfolded, the model identifies statically the processor schedule. The model therefore is useful for determining the throughput and latency of systems with heterogeneous processors. The applicability of the model is demonstrated using a space surveillance algorithm.

Leathrum, J. F.↗

DOE OSTI2022

Generating and Analyzing Program Call Graphs using Ontology

Call graph or caller-callee relationships have been used for various kinds of static program analysis, performance analysis and profiling, and for program safety or security analysis such as detecting anomalies of program execution or code injection attacks. However, different tools generate call graphs in different formats, which prevents efficient reuse of call graph results. In this paper, we present an approach of using ontology and resource description framework (RDF) to create knowledge graphs for specifying call graphs to facilitate the construction of full-fledged and complex call graphs of computer programs, realizing more interoperable and scalable program analyses than conventional approaches. We create a formal ontology-based specification of call graph information to capture concepts and properties of both static and dynamic call graphs so different tools can collaboratively contribute to more comprehensive analysis results. Our experiments show that ontology enables merging of call graphs generated from different tools and flexible queries using a standard query interface. Index Terms—Callgraph, ontology, knowl

Dorta, E.↗

NASA NTRS1980

Rolling flow wind tunnel tests of F-18 aircraft

The lateral directional characteristics of an F-18 aircraft was investigated. Aerodynamic derivatives associated with pure roll rate, or the 'p' derivatives were obtained. The model is described and the procedures used to obtain and correct the data, and a graphical presentation of the results are presented. These results include graphs of the lateral directional static stability derivatives versus angle of attack, and the lateral directional force and moment coefficients versus nondimensional roll rate. Results are presented for several configurations including complete, complete without vertical tails, complete without horizontal tails, fuselage wing and fuselage alone. Each of these configuations was tested with and without wing leading edge extensions. The basic control surfaces were deflected and the results were investigated.

Lutze, F. H.↗

NASA NTRS1986

An experimental study of the lift, drag and static longitudinal stability for a three lifting surface configuration

The experimental procedure and aerodynamic force and moment measurements for wind tunnel testing of the three lifting surface configuration (TLC) are described. The influence of nonelliptical lift distributions on lift, drag, and static longitudinal stability are examined; graphs of the lift coefficient versus angle of attack, the pitching moment coefficient, drag coefficient, and lift to drag ratio versus lift coefficient are provided. The TLC data are compared with the conventional tail-aft configuration and the canard-wing configuration; it is concluded that the TLC has better lift and high-lift drag characteristics, lift to drag ratio, and zero-lift moments than the other two configurations. The effects of variations in forward and tail wind incidence angles, gap, stagger, and forward wind span on the drag, lift, longitudinal stability, and zero-lift moments of the configuration are studied.

Ostowari, C.↗

NASA NTRS2003

Mechanisms of Rotating Instability in Axial Compressors Investigated

Rotating instability is a phenomenon that occurs in the tip flow region of axial compressor stages during stable operation. It can be observed in highly staggered rotors with significant tip clearance and is strongest at high-load operating points where the characteristic levels off. In this condition, the single-stage fan under investigation radiates an audible, whistling tone, and wall pressure spectra in the vicinity of the rotor disk exhibit nonrotational components. The graph shows the spectrum of static pressure at a point on the endwall near the leading edge. A hump appears at roughly half of the blade passing frequency (BPF) and is characteristic of rotating instability. A computational model was developed at the NASA Glenn Research Center to investigate the mechanism behind this phenomenon. A three-dimensional steady Navier-Stokes code that has been successfully tested for a wide range of turbomachinery flows was modified to execute a time-accurate simulation of the full annulus of the compressor. At the inlet of the computational domain, the total pressure, total temperature, and two velocity components are specified. Since no unsteady measurements of static pressure or other flow variables were available downstream of the rotor, circumferentially averaged static pressure was specified on the shroud at the outlet of the computational domain. A three-dimensional view of the vortex from the numerical model is shown. Particle traces released near the leading edge tip have rolled up to illustrate the tip clearance vortex. Flow near the trailing edge is pushed forward by the axially reversed flow. It then interacts with the tip clearance flow and the incoming flow and results in the rotating instability vortex, the core of which is illustrated by total pressure shading on planes located successively downstream. The rotating instability vortex is formed periodically midway between the blades and moves toward the pressure side of the passage. The unsteady behavior of this vortex structure is the main mechanism of the rotating instability is shown. The numerical model can be used to detect any possible occurrence of rotating instability when the tip clearance increases during engine service.

Hah, Chunill↗

DOE OSTI2023

Reduced-Order Models of Static Power Grids based on Spectral Clustering

For large-scale interconnected power systems that cover large geographical areas, certain electrical studies are required so that appropriate decisions ensure system reliability and low cost. For such studies, it is often neither practical nor necessary to model in detail the entire power system, which is increasingly complex due to a more diverse range of grid assets to choose from in both short and long-term planning. The goal of this paper is to present a methodology to reduce the order of large-scale power networks based on spectral graph theory given that current methods for static network reduction are not scalable. A brief analysis of some spectral clustering properties to determine which graph Laplacian matrix should be used and why is included. The analysis shows that the utilization of the normalized graph Laplacian is more advantageous for clustering purposes. Techniques are proposed to approximate cost functions for the aggregated generators. This is done via linear regression. The reduced-order model obtained with the proposed methodology has an accuracy above 94% and solves the scalability issue commonly present in other reduction methods. If the utilization of the reduced-order model is either constrained to load levels above mid-peak demand, or cost functions of aggregated units are approximated via a piecewise quadratic approach, then the error distribution is in the order of 10^-3. .

Baquedano-Aguilar, Mario D.↗