Search NASA⌕ Search

SEARCH · Search NASA

Results for “graph optimization”

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 523 records · Page 29

Attack detection and strategy optimization in game-theoretic trust models

Trust in a microelectronics-based systems can be characterized as the level of confidence that the system is free of subversive alterations inserted by a malicious adversary during system development. Outkin et al. recently developed GPLADD, a game-theoretic framework that enables trust analysis through a set of mathematical models that represent multi-step attack graphs and contention between system attackers and defenders. This paper extends GPLADD to include detection of attacks on development processes and defender decision processes that occur in response to detection events. The paper provides mathematical details for implementing attack detection and demonstrates the models on an example system. The authors further demonstrate how optimal defender strategies vary when solution concepts and objective functions are modified.

97 MATHEMATICS AND COMPUTING↗

Scalable training of graph convolutional neural networks for fast and accurate predictions of HOMO-LUMO gap in molecules

Abstract Graph Convolutional Neural Network (GCNN) is a popular class of deep learning (DL) models in material science to predict material properties from the graph representation of molecular structures. Training an accurate and comprehensive GCNN surrogate for molecular design requires large-scale graph datasets and is usually a time-consuming process. Recent advances in GPUs and distributed computing open a path to reduce the computational cost for GCNN training effectively. However, efficient utilization of high performance computing (HPC) resources for training requires simultaneously optimizing large-scale data management and scalable stochastic batched optimization techniques. In this work, we focus on building GCNN models on HPC systems to predict material properties of millions of molecules. We use HydraGNN, our in-house library for large-scale GCNN training, leveraging distributed data parallelism in PyTorch. We use ADIOS, a high-performance data management framework for efficient storage and reading of large molecular graph data. We perform parallel training on two open-source large-scale graph datasets to build a GCNN predictor for an important quantum property known as the HOMO-LUMO gap. We measure the scalability, accuracy, and convergence of our approach on two DOE supercomputers: the Summit supercomputer at the Oak Ridge Leadership Computing Facility (OLCF) and the Perlmutter system at the National Energy Research Scientific Computing Center (NERSC). We present our experimental results with HydraGNN showing (i) reduction of data loading time up to 4.2 times compared with a conventional method and (ii) linear scaling performance for training up to 1024 GPUs on both Summit and Perlmutter.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A differentiable approach to the maximum independent set problem using dataless neural networks

The success of machine learning solutions for reasoning about discrete structures has brought attention to its adoption within combinatorial optimization algorithms. Such approaches generally rely on supervised learning by leveraging datasets of the combinatorial structures of interest drawn from some distribution of problem instances. Reinforcement learning has also been employed to find such structures. Here, in this paper, we propose a different approach in that no data is required for training the neural networks that produce the solution. In this sense, what we present is not a machine learning solution, but rather one that is dependent on neural networks and where backpropagation is applied to a loss function defined by the structure of the neural network architecture as opposed to a training dataset. In particular, we reduce the popular combinatorial optimization problem of finding a maximum independent set to a neural network and employ a dataless training scheme to refine the parameters of the network such that those parameters yield the structure of interest. Additionally, we propose a universal graph reduction procedure to handle large-scale graphs. The reduction exploits community detection for graph partitioning and is applicable to any graph type and/or density. Experimental results on both real and synthetic graphs demonstrate that our proposed method performs on par or outperforms state-of-the-art learning-based methods in terms of the size of the found set without requiring any training data.

97 MATHEMATICS AND COMPUTING↗

Optimization Testbed Cometboards Extended into Stochastic Domain

COMparative Evaluation Testbed of Optimization and Analysis Routines for the Design of Structures (CometBoards) is a multidisciplinary design optimization software. It was originally developed for deterministic calculation. It has now been extended into the stochastic domain for structural design problems. For deterministic problems, CometBoards is introduced through its subproblem solution strategy as well as the approximation concept in optimization. In the stochastic domain, a design is formulated as a function of the risk or reliability. Optimum solution including the weight of a structure, is also obtained as a function of reliability. Weight versus reliability traced out an inverted-S-shaped graph. The center of the graph corresponded to 50 percent probability of success, or one failure in two samples. A heavy design with weight approaching infinity could be produced for a near-zero rate of failure that corresponded to unity for reliability. Weight can be reduced to a small value for the most failure-prone design with a compromised reliability approaching zero. The stochastic design optimization (SDO) capability for an industrial problem was obtained by combining three codes: MSC/Nastran code was the deterministic analysis tool, fast probabilistic integrator, or the FPI module of the NESSUS software, was the probabilistic calculator, and CometBoards became the optimizer. The SDO capability requires a finite element structural model, a material model, a load model, and a design model. The stochastic optimization concept is illustrated considering an academic example and a real-life airframe component made of metallic and composite materials.

Patnaik, Surya N.↗

GraphReduce: Processing Large-Scale Graphs on Accelerator-Based Systems

Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device’s internal memory capacity. GraphReduce adopts a combination of edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and device.

Architecture optimization, reuse, performance, ene↗

NCBI’s Virus Discovery Codeathon: Building “FIVE” —The Federated Index of Viral Experiments API Index

Viruses represent important test cases for data federation due to their genome size and the rapid increase in sequence data in publicly available databases. However, some consequences of previously decentralized (unfederated) data are lack of consensus or comparisons between feature annotations. Unifying or displaying alternative annotations should be a priority both for communities with robust entry representation and for nascent communities with burgeoning data sources. To this end, during this three-day continuation of the Virus Hunting Toolkit codeathon series (VHT-2), a new integrated and federated viral index was elaborated. This Federated Index of Viral Experiments (FIVE) integrates pre-existing and novel functional and taxonomy annotations and virus–host pairings. Variability in the context of viral genomic diversity is often overlooked in virus databases. As a proof-of-concept, FIVE was the first attempt to include viral genome variation for HIV, the most well-studied human pathogen, through viral genome diversity graphs. As per the publication of this manuscript, FIVE is the first implementation of a virus-specific federated index of such scope. FIVE is coded in BigQuery for optimal access of large quantities of data and is publicly accessible. Many projects of database or index federation fail to provide easier alternatives to access or query information. To this end, a Python API query system was developed to enhance the accessibility of FIVE.

59 BASIC BIOLOGICAL SCIENCES↗

Optimizing Traffic Signal Control to Enhance Transportation Efficiency and Maximize Pedestrian Benefits in the Road Network

Increasing urban mobility requirements demand efficient transportation system strategies for both vehicular and pedestrian movement. This study enhances the Decentralized Graph-based Multi-Agent Reinforcement Learning (DGMARL) approach, originally tailored for vehicular traffic signal timing, to incorporate pedestrian traffic dynamics. The improved algorithm considers crucial metrics such as Eco_PI, assesses vehicle fuel consumption by factoring in stops and delays, and addresses pedestrian waiting time, crucial for system efficiency while acknowledging driver waiting time impact. Utilizing Digital Twin simulation along the MLK Smart Corridor in Chattanooga, Tennessee, the algorithm's performance is compared for various pedestrian control scenarios. To evaluate the effectiveness of DGMARL, this study compared DGMARL-enabled signal management with automated pedestrian traffic detection and an actuated signal management system (real-word baseline) with pedestrian recall, which predetermingly enforces a pedestrian phase every cycle. Findings indicate substantial improvements with DGMARL, showing a 28.29% enhancement in vehicle Eco_PI, a 60.55 % reduction in pedestrian waiting time, and a 55.74% decrease in driver stop delay, on average, compared to the baseline actuated signal timing plan.

Kumarasamy, Vijayalakshmi K [The University of Ten↗

Distributed-memory Parallel Algorithms for Matching and Coloring

Graph matching and coloring constitute two fundamental classes of combinatorial problems having numerous established as well as emerging applications in computational science and engineering, high-performance computing, and informatics. We provide a snapshot of an on-going work on the design and implementation of new highly-scalable distributed-memory parallel algorithms for two prototypical problems from these classes, edge-weighted matching and distance-1 vertex coloring. Graph algorithms in general have low concurrency and poor data locality, making it challenging to achieve scalability on massively parallel machines. We overcome this challenge by employing a variety of techniques, including approximation, speculation and iteration, optimized communication, and randomization, in concert. We present preliminary results on weak and strong scalability studies conducted on an IBM Blue Gene/P machine employing up to tens of thousands of processors. The results show that the algorithms hold strong potential for computing at petascale.

Combinatorial scientific computing↗

CRNT4SBML: a Python package for the detection of bistability in biochemical reaction networks

Motivation: Signaling pathways capable of switching between two states are ubiquitous within living organisms. They provide the cells with the means to produce reversible or irreversible decisions. Switchlike behavior of biological systems is realized through biochemical reaction networks capable of having two or more distinct steady states which are dependent on initial conditions. Investigation of whether a certain signaling pathway can confer bistability involves a substantial amount of hypothesis testing. The cost of direct experimental testing can be prohibitive. Therefore, constraining the hypothesis space is highly bene?cial. One such methodology is based on Chemical Reaction Network Theory which uses computational techniques to rule out pathways that are not capable of bistability regardless of kinetic constant values and molecule concentrations. Although useful, these methods are complicated from both pureandcomputationalmathematicsperspectives.Thus,theiradoptionisverylimitedamongstbiologists. Results: We brought Chemical Reaction Network Theory approaches closer to experimental biologists by automating all the necessary steps in CRNT4SMBL. The input is based on SBML format which is the community standard for biological pathway communication. The tool parses SBML and derives C-graph representations of the biological pathway with mass action kinetics. Next steps involve an ef?cient search for potential saddle-node bifurcation points using an optimization technique. This type of bifurcation is important as it has the potential of acting as a switching point between two steady states. Finally, if any bifurcation points are present, numerical continuation analysis extends the equilibria branches for generating the diagram. Presence of an S-shaped bifurcation diagram indicates that the pathway acts as a bistable switch for the given optimization parameters. Availability: CRNT4SBML is available via the Python Package Index. The documentation can be found at https://crnt4sbml.readthedocs.io. CRNT4SBML is licensed under the Apache Software License 2.0. Contact: vladislav.petyuk@pnnl.gov

Reyes, Brandon C.↗

Heterogeneous Data Fusion via Space Alignment Using Nonmetric Multidimensional Scaling

Heterogeneous data sets are typically represented in different feature spaces, making it difficult to analyze relationships spanning different data sets even when they are semantically related. Data fusion via space alignment can remedy this task by integrating multiple data sets lying in different spaces into one common space. Given a set of reference correspondence data that share the same semantic meaning across different spaces, space alignment attempts to place the corresponding reference data as close together as possible, and accordingly, the entire data are aligned in a common space. Space alignment involves optimizing two potentially conflicting criteria: minimum deformation of the original relationships and maximum alignment between the different spaces. To solve this problem, we provide a novel graph embedding framework for space alignment, which converts each data set into a graph and assigns zero distance between reference correspondence pairs resulting in a single graph. We propose a graph embedding method for fusion based on nonmetric multidimensional scaling (MDS). Its criteria using the rank order rather than the distance allows nonmetric MDS to effectively handle both deformation and alignment. Experiments using parallel data sets demonstrate that our approach works well in comparison to existing methods such as constrained Laplacian eigenmaps, Procrustes analysis, and tensor decomposition. We also present standard cross-domain information retrieval tests as well as interesting visualization examples using space alignment.

Alignment↗

Efficient Parallelization of Irregular Applications on GPU Architectures

With the enlarging computation capacity of general Graphics Processing Units (GPUs), leveraging GPUs to accelerate parallel applications has become a critical topic in academia and industry. However, a wide range of irregular applications with the computation-/memory-intensive nature cannot easily achieve high GPU utilization. The challenges mainly involve the following aspects: first, data dependence leads to coarse-grained kernel and inefficient parallelism; second, heavy GPU memory usage may cause frequent memory evictions and extra overhead of I/O; third, specific computation patterns produce memory redundancies; last, workload balance and data reusability conjunctly benefit the overall performance, but there may exist a dynamic trade-off between them. Targeting these challenges, this dissertation proposes multiple optimizations to accelerate two real-world applications: many-body correlation functions to simulate nuclear physics in a large-scale scientific system; the other is the eALS-based matrix factorization recommendation system. To accelerate the calculations of many-body correlation functions, this dissertation presents three frameworks in GPU memory management and multi-GPU scheduling. Firstly, an optimized systematic GPU memory management framework, MemHC, utilizes a series of new memory reduction designs in GPU memory allocation, CPU/GPU communications, and GPU memory oversubscription. Secondly, an enhanced multi-GPU scheduling framework, MICCO, particularly by taking both data dimension (e.g., data reuse and data eviction) and computation dimension into account. MICCO designs a heuristic scheduling algorithm and a machine learning-based regression model to generate the optimal settings of a proposed new concept to manage the trade-off. Thirdly, a locality-aware multi-GPU scheduling framework. This scheduler leverages pipeline batch generation with a looking-ahead strategy by building local dependency graphs for memory transfer reduction and better data reuse, achieving up to 79.92% memory cost reduction and 1.67x speedup. To parallelize the eALS-based recommendation system, this dissertation proposes an efficient CPU/GPU heterogeneous recommendation system, HEALS. HEALS employs newly designed architecture-adaptive data formats to achieve load balance and good data locality on CPU and GPU. To mitigate the data dependence, HEALS presents a CPU/GPU collaboration model for both task parallelism and data parallelism with multiple kernel computation optimizations. In summary, this dissertation efficiently accelerates two typical irregular applications on GPUs by building four frameworks, including CPU/GPU collaboration, GPU memory management, and multi-GPU scheduling.

Wang, Qihan↗

Learning nuclear cross sections across the chart of nuclides with graph neural networks

We explore the use of deep learning techniques to learn how nuclear cross sections change as we add or remove protons and neutrons. As a proof of principle, we focus on the neutron-induced reactions in the fast energy regime. Our approach follows a two-stage learning framework. First, we apply representation learning to encode cross section data into a latent space using either variational autoencoders (VAEs) or implicit neural representations (INRs). Then, we train graph neural networks (GNNs) on the resulting embeddings to predict missing values across the nuclear chart by leveraging the topological structure of neighboring isotopes. We demonstrate accurate cross section predictions within a 9 × 9 block of missing nuclei. We also find that the optimal GNN training strategy depends on the type of latent representation used, with VAE embeddings performing best under end-to-end optimization in the original space, while INR embeddings achieve better results when the GNN is trained only in the latent space. Furthermore, using clustering algorithms, we map groups of latent vectors into regions of the nuclear chart and show that VAEs and INRs can discover some of the neutron magic numbers. These findings suggest that deep-learning models based on the representation encoding of cross sections combined with graph neural networks hold significant potential in augmenting nuclear theory models, e.g., by providing reliable estimates of covariances of cross sections, including cross-material covariances.

Machine learning↗

Distributed Louvain Algorithm for Graph Community Detection

In most real-world networks, the nodes/vertices tend to be organized into tightly-knit modules known as communities or clusters, such that nodes within a community are more likely to be “related” to one another than they are to the rest of the network. The goodness of partitioning into communities is typically measured using a well known measure called modularity. However, modularity optimization is an NP-complete problem. In 2008, Blondel, et al. introduced a multi-phase, iterative heuristic for modularity optimization, called the Louvain method. Owing to its speed and ability to yield high quality communities, the Louvain method continues to be one of the most widely used tools for serial community detection. In this paper, we present the design of a distributed memory implementation of the Louvain algorithm for parallel community detection. Our approach begins with an arbitrarily partitioned distributed graph input, and employs several heuristics to speedup the computation of the different steps of the Louvain algorithm. We evaluate our implementation and its different variants using real-world networks from various application domains (including internet, biology, social networks). Our MPI+OpenMP implementation yields about 7x speedup (on 4K processes) for soc-friendster network (1.8B edges) over a state-of-the-art shared-memory multicore implementation (on 64 threads), without compromising output quality. Furthermore, our distributed implementation was able to process a larger graph (uk-2007; 3.3B edges) in 32 seconds on 1K cores (64 nodes) of NERSC Cori, when the state-of-the-art shared-memory implementation failed to run due to insufficient memory on a single Cori node containing 128 GB.

High performance computing, graph algorithms, Dist↗

IRIS-GNN: Leveraging Graph Neural Networks for Scheduling on Truly Heterogeneous Runtime Systems

The diversity of accelerators in computer systems poses significant challenges for software developers, such as managing vendor-specific compiler toolchains, code fragmentation requiring different kernel implementations, and performance portability issues. To address these, the Intelligent Runtime System (IRIS) was developed. IRIS works across various systems, from smartphones to supercomputers, enabling automatic performance scaling based on available accelerators. It introduces abstract tasks for seamless execution transitions between accelerators while ensuring memory consistency and task dependencies. Although IRIS simplifies system details, optimal dynamic scheduling still requires user input to understand workload structures. To address this, we introduce a new scheduling policy for IRIS, termed IRIS-GNN, which is the first IRIS hybrid policy that operates in conjunction with the dynamic policies. This policy employs a Graph-Neural Network (GNN) to conduct Graph Classification of any task graphs submitted to IRIS. This GNN analyzes the structure and attributes of the task graph, categorizing it as either locality, concurrency, or mixed. This classification subsequently guides the selection of the dynamic policy used by IRIS. We provide a comparison of the performance of IRIS-GNN against the complete spectrum of IRIS’s dynamic policies, assess the overhead introduced by the GNN within this scheduling framework, and ultimately explore its practical application in real-world scenarios.

Johnston, Beau↗

Conflict Detection in Open RAN with Recurrent Neural Networks Using Geometric Manifolds

Allowing third-party applications on Radio Access Network (RAN) Intelligent Controllers (RICs) within the OpenRAN (O-RAN) framework introduces conflicting interactions that are often difficult to detect in advance. These conflicts, occurring between third-party applications in the Near RealTime RIC (Near-RT RIC), known as xApps, can lead to performance degradation and instability in O-RAN if not identified early. Existing conflict detection and mitigation solutions in the literature assume that the conflicts are known beforehand, which is not always accurate due to the complex and often hidden relationships between control parameters and Key Performance Indicators (KPIs). In this paper, we propose a novel Recurrent Neural Network (RNN) to detect both known and unknown conflicts in O-RAN xApps as specified in the O-RAN standards. We model the xApps, control parameters, and KPIs with nodes and edges to create graph structures and use the hidden nonEuclidean geometric properties of the Riemannian manifold to train the RNN model. The performance of this proposed model is validated using evaluation metrics and compared with benchmarks. Results demonstrate that the proposed RNN model, leveraging Riemannian geometric properties, can achieve 100% of the F1-score provided by an optimal solution in just 20 iterations.

5G↗

Conflict Detection in Open RAN with Recurrent Neural Networks Using Geometric Manifolds

Allowing third-party applications on Radio Access Network (RAN) Intelligent Controllers (RICs) within the OpenRAN (O-RAN) framework introduces conflicting interactions that are often difficult to detect in advance. These conflicts, occurring between third-party applications in the Near RealTime RIC (Near-RT RIC), known as xApps, can lead to performance degradation and instability in O-RAN if not identified early. Existing conflict detection and mitigation solutions in the literature assume that the conflicts are known beforehand, which is not always accurate due to the complex and often hidden relationships between control parameters and Key Performance Indicators (KPIs). In this paper, we propose a novel Recurrent Neural Network (RNN) to detect both known and unknown conflicts in O-RAN xApps as specified in the O-RAN standards. We model the xApps, control parameters, and KPIs with nodes and edges to create graph structures and use the hidden nonEuclidean geometric properties of the Riemannian manifold to train the RNN model. The performance of this proposed model is validated using evaluation metrics and compared with benchmarks. Results demonstrate that the proposed RNN model, leveraging Riemannian geometric properties, can achieve 100% of the F1-score provided by an optimal solution in just 20 iterations.

5G↗

GraphReduce: Large-Scale Graph Analytics on Accelerator-Based HPC Systems

Recent work on real-world graph analytics has sought to leverage the massive amount of parallelism offered by GPU devices, but challenges remain due to the inherent irregularity of graph algorithms and limitations in GPU-resident memory for storing large graphs. We present GraphReduce, a highly efficient and scalable GPU-based framework that operates on graphs that exceed the device’s internal memory capacity. GraphReduce adopts a combination of both edge- and vertex-centric implementations of the Gather-Apply-Scatter programming model and operates on multiple asynchronous GPU streams to fully exploit the high degrees of parallelism in GPUs with efficient graph data movement between the host and the device.

Architecture optimization, reuse, performance, ene↗

AutoCheck: Automatically Identifying Variables for Checkpointing by Data Dependency Analysis

Checkpoint/Restart (C/R) has been widely deployed in numerous HPC systems, Clouds, and industrial data centers, which are typically operated by system engineers. Nevertheless, there is no existing approach that helps system engineers without domain expertise and domain scientists without system fault tolerance knowledge identify those critical variables accounted for correct application execution restoration in a failure for C/R. To address this problem, we propose an analytical model and a tool (AutoCheck) that can automatically identify critical variables to checkpoint for C/R. AutoCheck relies on first, analytically tracking and optimizing data dependency between variables and other application execution state, and second, a set of heuristics that identify critical variables for checkpointing from the refined data dependency graph (DDG). AutoCheck allows programmers to pinpoint critical variables to checkpoint quickly within a few minutes. We evaluate AutoCheck on 13 representative HPC benchmarks, demonstrating that AutoCheck can efficiently identify correct critical variables to checkpoint.

HPC↗