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 379 records · Page 21

SWAP-Assembler 2: Optimization of De Novo Genome Assembler at Large Scale

In this paper, we analyze and optimize the most time-consuming steps of the SWAP-Assembler, a parallel genome assembler, so that it can scale to a large number of cores for huge genomes with the size of sequencing data ranging from terabyes to petabytes. According to the performance analysis results, the most time-consuming steps are input parallelization, k-mer graph construction, and graph simplification (edge merging). For the input parallelization, the input data is divided into virtual fragments with nearly equal size, and the start position and end position of each fragment are automatically separated at the beginning of the reads. In k-mer graph construction, in order to improve the communication efficiency, the message size is kept constant between any two processes by proportionally increasing the number of nucleotides to the number of processes in the input parallelization step for each round. The memory usage is also decreased because only a small part of the input data is processed in each round. With graph simplification, the communication protocol reduces the number of communication loops from four to two loops and decreases the idle communication time. The optimized assembler is denoted as SWAP-Assembler 2 (SWAP2). In our experiments using a 1000 Genomes project dataset of 4 terabytes (the largest dataset ever used for assembling) on the supercomputer Mira, the results show that SWAP2 scales to 131,072 cores with an efficiency of 40%. We also compared our work with both the HipMER assembler and the SWAP-Assembler. On the Yanhuang dataset of 300 gigabytes, SWAP2 shows a 3X speedup and 4X better scalability compared with the HipMer assembler and is 45 times faster than the SWAP-Assembler. The SWAP2 software is available at https://sourceforge.net/projects/swapassembler.

Meng, Jintao↗

Sampling frequency thresholds for the quantum advantage of the quantum approximate optimization algorithm

We compare the performance of the Quantum Approximate Optimization Algorithm (QAOA) with state-of-the-art classical solvers Gurobi and MQLib to solve the MaxCut problem on 3-regular graphs. We identify the minimum noiseless sampling frequency and depth p required for a quantum device to outperform classical algorithms. There is potential for quantum advantage on hundreds of qubits and moderate depth with a sampling frequency of 10 kHz. We observe, however, that classical heuristic solvers are capable of producing high-quality approximate solutions in linear time complexity. In order to match this quality for large graph sizes N, a quantum device must support depth p > 11. Additionally, multi-shot QAOA is not efficient on large graphs, indicating that QAOA p ≤ 11 does not scale with N. These results limit achieving quantum advantage for QAOA MaxCut on 3-regular graphs. Other problems, such as different graphs, weighted MaxCut, and 3-SAT, may be better suited for achieving quantum advantage on near-term quantum devices.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Algebraic Algorithms for Betweenness and Percolation Centrality

Abstract In this paper, we explored different ways to write the algebraic version of betweenness centrality algorithm. Particularly, we focused on Brandes' algorithm. We aimed for algebraic betweenness centrality that can be parallelized easily. We proposed 3-tuple geodetic semiring as an extension to the usual geodetic semiring with 2-tuples. Using the 3-tuple geodetic semiring, Dijkstra's and Brandes' algorithm, we wrote more concise and general algebraic betweenness centrality (ABC) algorithm which is valid for weighted and directed graphs. We also proposed an alternative version of ABC using the usual geodetic semiring with 2-tuple where we used a simple way to construct shortest path tree after computing shortest path distances in the usual geodetic semiring. This allows us to avoid computational complexity of ABC implementation using 3-tuple geodetic semiring. We used numba to optimize and parallelize ABC. We evaluated the performance of ABC using 2-tuple geodetic semiring as compared to NetworkX, a common python package for graph algorithms. We did scalability experiments on parallel ABC and showed its total speedup. We also showed that with small modification, ABC can be adapted to algebraicly compute other centrality measures such as percolation centrality.

97 MATHEMATICS AND COMPUTING↗

Optimizing Irregular Applications for Energy and Performance on the Tilera Many-core Architecture

Optimizing applications simultaneously for energy and performance is a complex problem. High performance, parallel, irregular applications are notoriously hard to optimize due to their data-dependent memory accesses, lack of structured locality and complex data structures and code patterns. Irregular kernels are growing in importance in applications such as machine learning, graph analytics and combinatorial scientific computing. Performance- and energy-efficient implementation of these kernels on modern, energy efficient, multicore and many-core platforms is therefore an important and challenging problem. We present results from optimizing two irregular applications { the Louvain method for community detection (Grappolo), and high-performance conjugate gradient (HPCCG) { on the Tilera many-core system. We have significantly extended MIT's OpenTuner auto-tuning framework to conduct a detailed study of platform-independent and platform-specific optimizations to improve performance as well as reduce total energy consumption. We explore the optimization design space along three dimensions: memory layout schemes, compiler-based code transformations, and optimization of parallel loop schedules. Using auto-tuning, we demonstrate whole node energy savings of up to 41% relative to a baseline instantiation, and up to 31% relative to manually optimized variants.

irregular applications, energy opttimization, many↗

Real-Time Identifiability of Power Distribution Network Topologies With Limited Monitoring

Recovering the distribution grid topology in real time is essential to perform several distribution system operator (DSO) functions. DSOs often do not have any direct monitoring of switch statuses to track reconfiguration. At the same time, installing real-time meters at a large number of buses is challenging due to the cost of endowing every metered bus with a real-time communication channel. The goal of this letter is to develop a meter placement strategy allowing DSOs to deploy only few real-time meters. After casting the topology recovery task as an optimization problem, a meter placement strategy ensuring unique recovery of the true topology is devised. A graph-theoretical approach is pursued to partition the grid into connected portions called observable islands . The proposed strategy then simply requires installing a meter in the path between every pair of boundary nodes , i.e., ends of edges connecting two different islands. Under some ideal assumptions, this placement strategy ensures unique recovery of the topology. The approach is also validated through numerical simulations under realistic scenarios using a standard IEEE benchmark feeder.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Compare linear-system solver and preconditioner stacks with emphasis on GPU performance and propose phase-2 NGP solver development pathway

The goal of the ExaWind project is to enable predictive simulations of wind farms comprised of many megawatt-scale turbines situated in complex terrain. Predictive simulations will require computational fluid dynamics (CFD) simulations for which the mesh resolves the geometry of the turbines and captures the rotation and large deflections of blades. Whereas such simulations for a single turbine are arguably petascale class, multi-turbine wind farm simulations will require exascale-class resources. The primary physics codes in the ExaWind project are Nalu-Wind, which is an unstructured-grid solver for the acoustically incompressible Navier-Stokes equations, and OpenFAST, which is a whole-turbine simulation code. The Nalu-Wind model consists of the mass-continuity Poisson-type equation for pressure and a momentum equation for the velocity. For such modeling approaches, simulation times are dominated by linear-system setup and solution for the continuity and momentum systems. For the ExaWind challenge problem, the moving meshes greatly affect overall solver costs as reinitialization of matrices and recomputation of preconditioners is required at every time step. In this report we evaluated GPU-performance baselines for the linear solvers in the Trilinos and hypre solver stacks using two representative Nalu-Wind simulations: an atmospheric boundary layer precursor simulation on a structured mesh, and a fixed-wing simulation using unstructured overset meshes. Both strong-scaling and weak-scaling experiments were conducted on the OLCF supercomputer Summit and similar proxy clusters. We focused on the performance of multi-threaded Gauss-Seidel and two-stage Gauss-Seidel that are extensions of classical Gauss-Seidel; of one-reduce GMRES, a communication-reducing variant of the Krylov GMRES; and algebraic multigrid methods that incorporate the afore-mentioned methods. The team has established that AMG methods are capable of solving linear systems arising from the fixed-wing overset meshes on CPU, a critical intermediate result for ExaWind FY20 Q3 and Q4 milestones. For the fixed-wing strong-scaling study (model with 3M grid-points), the team identified that Nalu-Wind simulations with the new Trilinos and hypre solvers scale to modest GPU counts, maintaining above 70% efficiency up to 6 GPUs. However, there still remain significant bottlenecks to performance: matrix assembly (hypre), AMG setup (hypre and Trilinos) In the weak-scaling experiments (going from 0.4M to 211M gridpoints), it's shown that the solver apply phases are faster on GPUs, but that Nalu-Wind simulation times grow, primarily due to the multigrid-setup process. Finally, based on the report outcomes, we propose a linear solver path-forward for the remainder of the ExaWind project. Near term, the NREL team will continue their work on GPU-based linear-system assembly. They will also investigate how the use of alternatives to the NVIDIA UVM (unified virtual memory) paradigm affects performance. Longer term, the NREL team will evaluate algorithmic performance on other types of accelerators and merge their improvements back to the main hypre repository branch. Near term, the Trilinos team will address performance bottlenecks identified in this milestone, such as implementing a GPU-based segregated momentum solve and reusing matrix graphs across linear-system assembly phases. Longer term, the Trilinos team will do detailed analysis and optimization of multigrid setup.

17 WIND ENERGY↗

Design and optimization of stepped austempered ductile iron using characterization techniques

Conventional characterization techniques such as dilatometry, X-ray diffraction and metallography were used to select and optimize temperatures and times for conventional and stepped austempering. Austenitization and conventional austempering time was selected when the dilatometry graphs showed a constant expansion value. A special heat color-etching technique was applied to distinguish between the untransformed austenite and high carbon stabilized austenite which had formed during the treatments. Finally, it was found that carbide precipitation was absent during the stepped austempering in contrast to conventional austempering, on which carbide evidence was found. - Highlights: • Dilatometry helped to establish austenitization and austempering parameters. • Untransformed austenite was present even for longer processing times. • Ausferrite formed during stepped austempering caused important reinforcement effect. • Carbide precipitation was absent during stepped treatment.

36 MATERIALS SCIENCE↗

Graph Algorithms in PGAS: Chapel and UPC++

The Partitioned Global Address Space (PGAS) programming model can be implemented «««< HEAD either with programming language features or with runtime library APIs, each implementation favoring different aspects (e.g., productivity, abstraction, flexibility, or performance). Certain language and runtime features, such as collectives, explicit and asynchronous communication primitives, and constructs facilitating overlap of communication and computation (such as futures and conjoined futures) can enable better performance and scaling for irregular applications, in particular for distributed graph analytics. We compare graph algorithms in one of each of these environments: the Chapel PGAS programming language and the UPC++ PGAS runtime library. We implement algorithms for breadth-first search and triangle counting graph kernels in both environments. We discuss the code in each of the environments, and compile performance data on a Cray Aries and an Infiniband platform. Our results show that the library-based approach of UPC++ currently provides strong performance while Chapel provides a high-level abstraction that, harder to optimize, still provides comparable performance. We argue that various language primitives and runtime library features should be considered in conjunction with each other to design future graph analytics libraries so as to incorporate the best of both worlds.

Distributed Algorithms, Graph algorithms, HPC↗

Expediting field-effect transistor chemical sensor design with neuromorphic spiking graph neural networks

Improving the sensitive and selective detection of analytes in a variety of applications requires accelerating the rational design of field-effect transistor (FET) chemical sensors. Achieving high-performance detection relies on identifying optimal probe materials that can effectively interact with target analytes, a process traditionally driven by chemical intuition and time-consuming trial-and-error methods. To address the difficulties in probe screening for FET sensor development, this work presents a methodology that combines neuromorphic machine learning (ML) architectures, specifically a hybrid spiking graph neural network (SGNN), with an enriched dataset of physicochemical properties through semi-automated data extraction using large language models. Achieving a classification accuracy of 0.89 in predicting sensor sensitivity categories, the SGNN model outperformed traditional ML techniques by leveraging its ability to capture both global physicochemical properties and sparse topological features through a hybrid modeling framework. Next-generation sensor design was informed by the actionable insights into the connections between material properties and sensing performance offered by the SGNN framework. Through virtual screening for the detection of per- and polyfluoroalkyl substances (PFAS) as a use case, the effectiveness of the SGNN model was further validated. Density functional theory simulations confirmed graphene as a promising active material for PFAS detection as suggested by the SGNN framework. By bridging gaps in predictive modeling and data availability, this integrated approach provides a strong foundation for accelerating advancements in FET sensor design and innovation.

Ferreira, Rodrigo Pires [Univ. of Chicago, IL (Uni↗

Scalable Graph Analytics and HPC Operational Enhancement: Parallel Computing and ML/DL Innovations

Parallel computing plays a pivotal role in the efficient processing of large-scale graphs. Complex network analysis stands as a capti- vating research frontier, holding promise across diverse scientific domains such as sociology, biology, online media, and recommenda- tion systems. In this era, Machine Learning (ML) and Deep Learning (DL) have emerged as indispensable tools, underpinning remarkable technological achievements. Within this dynamic landscape, my research revolves around advancing parallel algorithms tailored for large-scale graph operations. To achieve this, I harness the power of cutting-edge technologies including OpenMP, MPI, HIP, and CUDA, on the High-Performance Computing (HPC) platforms to unlock optimal performance. I also apply ML/DL techniques to HPC operational data, to streamline the monitoring and maintenance of supercomputers, alleviating the complexities associated with their upkeep and enhancing user support. My research echoes the syn- ergy between parallel computing, large-scale graph analysis, and ML/DL, improving computational efficiency and user experience.

Sattar, Naw Safrin↗

A Multithreaded Algorithm for Network Alignment Via Approximate Matching

Network alignment is an optimization problem to find the best one-to-one map between the vertices of a pair of graphs that overlaps in as many edges as possible. It is a relaxation of the graph isomorphism problem and is closely related to the subgraph isomorphism problem. The best current approaches are entirely heuristic, and are iterative in nature. They generate real-valued heuristic approximations that must be rounded to find integer solutions. This rounding requires solving a bipartite maximum weight matching problem at each step in order to avoid missing high quality solutions. We investigate substituting a parallel, half-approximation for maximum weight matching instead of an exact computation. Our experiments show that the resulting difference in solution quality is negligible. We demonstrate almost a 20-fold speedup using 40 threads on an 8 processor Intel Xeon E7-8870 system (from 10 minutes to 36 seconds).

network alignment, matching, multicore↗

Enabling Scalable and Extensible Memory-mapped Datastores in Userspace

Exascale workloads are expected to incorporate data-intensive processing in close coordination with traditional physics simulations. These emerging scientific, data-analytics and machine learning applications need to access a wide variety of datastores in flat files and structured databases. Programmer productivity is greatly enhanced by mapping datastores into the application process's virtual memory space to provide a unified “in-memory” interface. Currently, memory mapping is provided by system software primarily designed for generality and reliability. However, scalability at high concurrency is a formidable challenge on exascale systems. Also, there is a need for extensibility to support new datastores potentially requiring HPC data transfer services. In this article, we present UMap , a scalable and extensible userspace service for memory-mapping datastores. Furthermore, through decoupled queue management, concurrency aware adaptation, and dynamic load balancing, UMap enables application performance to scale even at high concurrency. We evaluate UMap in data-intensive applications, including sorting, graph traversal, database operations, and metagenomic analytics. Our results show that UMap as a userspace service outperforms an optimized kernel-based service across a wide range of intra-node concurrency by 1.22-1.9 × . We performed two case studies to demonstrate UMap 's extensibility. First, a new datastore residing in remote memory is incorporated into UMap as an application-specific plugin. Second, we present a persistent memory allocator Metall built atop UMap for unified storage/memory.

97 MATHEMATICS AND COMPUTING↗

Runtime Systems for Energy Efficiency in Advanced Computing Systems

As heterogeneous systems become increasingly popular for both mobile and high-performance computing, conventional efficiency techniques such as dynamic voltage and frequency scaling (DVFS) fail to account for the tightly coupled and varied nature of systems on a chip (SoCs). In this work, we explore the impact of system unaware DVFS techniques on a mobile SoC under three benchmark suites: Chai, Rodinia, and Antutu. We then analyze performance trends across the suites to identify a set of consistent operating points that optimally balance power and performance across the system. The consistent operating points are then constructed into a dependency graph which can be leveraged to produce a more effective, SoC-wide governor.

97 MATHEMATICS AND COMPUTING↗

Cyber Network Design for Secondary Frequency Regulation: A Spectral Approach

We present a preliminary theoretical framework based on spectral graph theory that captures how the cyber topology of a distributed secondary frequency control scheme impacts the stability, optimality, and transient performance of our power system as a cyber-physical network. We show that a collection of polynomials defined in terms of the cyber and physical Laplacian eigenvalues encode information on the interplay between cyber and physical networks. It is demonstrated that to understand the impact of adding cyber connectivity, one should separate the low-damping and high-damping regimes. Although adding cyber connectivity always improves the performance for high-damping systems, it is not the case for low-damping scenarios. Based on the theoretical study, we discuss how a good cyber network should be designed. Our empirical study shows that for practical systems, the number of communication channels that is needed to achieve near-optimal performance is usually less than twice the number of buses.

24 POWER TRANSMISSION AND DISTRIBUTION↗

System and method for reasoning about the optimality of a configuration parameter of a distributed system

Embodiments provide a system and method for reasoning about the optimality of a configuration parameter of a distributed system. During operation, the system obtains a multi-layer graph for a system with a plurality of components, wherein the multi-layer graph comprises a configuration subgraph, a vulnerability subgraph, and a dependency subgraph. The system determines, based on the multi-layer graph, constraint relationships associated with configuration parameters for the components, wherein the constraint relationships include security constraints and functionality constraints. The system computes an unsatisfiable core which comprises a set of mutually incompatible constraints. The system resolves, based on a strategy and over multiple iterations, the unsatisfiable core by analyzing one pair of mutually incompatible constraints per a respective iteration, to obtain a new unsatisfiable core which comprises a smaller number of mutually incompatible constraints than the computed unsatisfiable core or a previously computed unsatisfiable core from a most recent iteration.

Soroush, Hamed↗

Applying graph partitioning methods in measurement-based dynamic load balancing

Load imbalance leads to an increasing waste of resources as an application is scaled to more and more processors. Achieving the best parallel efficiency for a program requires optimal load balancing which is a NP-hard problem. However, finding near-optimal solutions to this problem for complex computational science and engineering applications is becoming increasingly important. Charm++, a migratable objects based programming model, provides a measurement-based dynamic load balancing framework. This framework instruments and then migrates over-decomposed objects to balance computational load and communication at runtime. This paper explores the use of graph partitioning algorithms, traditionally used for partitioning physical domains/meshes, for measurement-based dynamic load balancing of parallel applications. In particular, we present repartitioning methods developed in a graph partitioning toolbox called SCOTCH that consider the previous mapping to minimize migration costs. We also discuss a new imbalance reduction algorithm for graphs with irregular load distributions. We compare several load balancing algorithms using microbenchmarks on Intrepid and Ranger and evaluate the effect of communication, number of cores and number of objects on the benefit achieved from load balancing. New algorithms developed in SCOTCH lead to better performance compared to the METIS partitioners for several cases, both in terms of the application execution time and fewer number of objects migrated.

97 MATHEMATICS AND COMPUTING↗

A spatial superstructure approach to the optimal design of modular processes and supply chains

Modularity is a design principle that aims to provide flexibility for spatio-temporal assembly/disassembly and reconfiguration of systems. This design principle can be applied to multiscale (hierarchical) manufacturing systems that connect units, processes, facilities, and entire supply chains. Designing modular systems is challenging because of the need to capture spatial interdependencies that arise between system components due to product exchange/transport between components and due to product transformation in such components. In this work, we propose an optimization framework to facilitate the design of modular manufacturing systems. Central to our approach is the concept of a spatial superstructure, which is a graph that captures all possible system configurations and interdependencies between components. The spatial superstructure is a generalization of the notion of a superstructure and of a p-graph used in process design, in that it encodes spatial (geographical) context of the system components. Here, we show that this generalization facilitates the simultaneous design and analysis of processes, facilities, and of supply chains. Our framework aims to select the system topology from the spatial superstructure that minimizes design cost and that maximizes design modularity. We show that this design problem can be cast as a mixed-integer, multi-objective optimization formulation. We demonstrate these capabilities using a case study arising in the design of a plastic waste upcycling supply chain.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

MDLoader: A Hybrid Model-Driven Data Loader for Distributed Graph Neural Network Training

Scalable data management is essential for processing large scientific dataset on HPC platforms for distributed deep learning. In-memory distributed storage is preferred for its speed, enabling rapid, random, and frequent data access required by stochastic optimizers. Processes use one-sided or collective communication to fetch remote data, with optimal performance depending on (i) dataset characteristics, (ii) training scale, and (iii) interconnection network. Empirical analysis shows collective communication excels with larger mini-batch sizes and/or fewer processes, whereas one-sided communication outperforms at larger scales. We propose MDLoader, a hybrid in-memory data loader for distributed graph neural network training. MDLoader features a model-driven performance estimator that dynamically selects between one-sided and collective communication at the beginning of training using Tree of Parzen Estimators (TPE). Evaluations on NERSC Perlmutter and OLCF Summit show MDLoader outperforms single-backend loaders by up to 2.83 × and predicts the suitable communication method with 96.3% (Perlmutter) and 94.3% (Summit) success rate.

Bae, Jonghyun↗