Search NASA⌕ Search

SEARCH · Search NASA

Results for “scalable performance”

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 109 records · Page 6

Synchronization-Avoiding Graph Algorithms

Because they were developed for optimal sequential complexity, classical graph algorithms as found in textbooks have strictly-defined orders of operations. Enforcing a prescribed order of operations, or even an approximate order, in a distributed memory setting requires significant amounts of synchronization, which in turn can severely limit scalability. As a result, new algorithms are typically required to achieve scalable performance, even for solving well-known graph problems. Yet, even in these cases, parallel graph algorithms are written according to parallel programming models that evolved for, e.g., scientific computing, and that still have inherent, and scalability-limiting, amounts of synchronization. In this paper we present a new approach to parallel graph algorithms: synchronization-avoiding algorithms. To eliminate synchronization and its associated overhead, synchronization-avoiding algorithms perform work in an unordered and fully asynchronous fashion in such a way that the result is constantly refined toward its final state. “Wasted” work is minimized by locally prioritizing tasks using problem-dependent task utility metrics. We classify algorithms for graph applications into two broad categories: algorithms with monotonic updates (which evince global synchronization) and algorithms with nonmonotonic updates (which evince vertex-centric synchronization). We apply our approach to both classes and develop novel, synchronization-avoiding algorithms for solving exemplar problems: SSSP and connected components for the former, graph coloring for the latter. We demonstrate that eliminating synchronization in conjunction with effective scheduling policies and optimizations in the runtime results in improved scalability for both classes of algorithms.

Firoz, Jesun S.↗

Scalable Work Stealing

Irregular and dynamic parallel applications pose significant challenges to achieving scalable performance on large-scale multicore clusters. These applications often require ongoing, dynamic load balancing in order to maintain efficiency. While effective at small scale, centralized load balancing schemes quickly become a bottleneck on large-scale clusters. Work stealing is a popular approach to distributed dynamic load balancing; however its performance on large-scale clusters is not well understood. Prior work on work stealing has largely focused on shared memory machines. In this work we investigate the design and scalability of work stealing on modern distributed memory systems. We demonstrate high efficiency and low overhead when scaling to 8,192 processors for three benchmark codes: a producer-consumer benchmark, the unbalanced tree search benchmark, and a multiresolution analysis kernel.

99 GENERAL AND MISCELLANEOUS↗

Toward scalable information processing with ultracold polar molecules in an electric field: A numerical investigation

We numerically investigate the possibilities of driving quantum algorithms with laser pulses in a register of ultracold NaCs polar molecules in a static electric field. We focus on the possibilities of performing scalable logical operations by considering circuits that involve intermolecular gates (implemented on adjacent interacting molecules) to enable the transfer of information from one molecule to another during conditional laser-driven population inversions. We study the implementation of an arithmetic operation (the addition of 0 or 1 on a binary digit and a carry in) which requires population inversions only and the Deutsch-Jozsa algorithm which requires a control of the phases. Under typical experimental conditions, our simulations show that high-fidelity logical operations involving several qubits can be performed in a time scale of a few hundreds of microseconds, opening promising perspectives for the manipulation of a large number of qubits in these systems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Porting Adaptive Ensemble Molecular Dynamics Workflows to the Summit Supercomputer

Molecular dynamics (MD) simulations must take very small (femtosecond) integration steps in simulation-time to avoid numerical errors. Efficient use of parallel programming models and accelerators in state-of-the art MD programs now is pushing Moore’s limit for time-per-MD step. As a result, directly simulating timescales beyond milliseconds will not be attainable directly, even at exascale. However, concepts from statistical physics can be used to combine many parallel simulations to provide information about longer timescales and to adequately sample the simulation space, while preserving details about the dynamics of the system. Implementing such an approach requires a workflow program that allows adaptable steering of task assignments based on extensive statistical analysis of intermediate results. Here we report the implementation of such an adaptable workflow program to drive simulations on the Summit IBM Power System AC922, a pre-exascale supercomputer at the Oak Ridge Leadership Computing Facility (OLCF). We compare to experiences on Titan, Summit’s predecessor, report the performance of the workflow and its components, and describe the porting process. We find that using a workflow program managed by a Mongo database can provide the fault tolerance, scalable performance, task dispatch rate, and reconfigurability required for robust and portable implementation of ensemble simulations such as are used in enhanced-sampling molecular dynamics. This type of workflow generator can also be used to provide adaptive steering of ensemble simulations for other applications in addition to MD.

Ossyra, John R.↗

Nek5000/RS performance on advanced GPU architectures

The authors explore performance scalability of the open-source thermal-fluids code, NekRS, on the U.S. Department of Energy's leadership computers, Crusher, Frontier, Summit, Perlmutter, and Polaris. Particular attention is given to analyzing performance and time-to-solution at the strong-scale limit for a target efficiency of 80%, which is typical for production runs on the DOE's high-performance computing systems. Several examples of anomalous behavior are also discussed and analyzed.

97 MATHEMATICS AND COMPUTING↗

Aces4: A Platform for Computational Chemistry Calculations with Extremely Large Block-Sparse Arrays

Aces4 is a parallel programming platform comprising a DSL for Computational Chemistry and its runtime system. It offers a convenient way to express parallelism together with extensive support for extremely large, possibly sparse, distributed arrays. It aids scientists in the creation of performant, scalable, massively parallel programs that can effectively take advantage of leadership class computing systems to address important scientific questions. Aces4 has enabled the development and implementation of new methods in electronic structure theory which are breaking new ground in their ability to perform highly accurate calculations on ever larger molecular systems. In this paper the design of Aces4, which is based on the the Super Instruction Architecture approach, is described. Experimental scaling results for Molecular Cluster Perturbation Theory, a new method enabled by Aces4, and CCSD, a widely used computational chemistry method are given.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Visualization on supercomputing platform level II ASC milestone (3537-1B) results from Sandia.

This report provides documentation for the completion of the Sandia portion of the ASC Level II Visualization on the platform milestone. This ASC Level II milestone is a joint milestone between Sandia National Laboratories and Los Alamos National Laboratories. This milestone contains functionality required for performing visualization directly on a supercomputing platform, which is necessary for peta-scale visualization. Sandia's contribution concerns in-situ visualization, running a visualization in tandem with a solver. Visualization and analysis of petascale data is limited by several factors which must be addressed as ACES delivers the Cielo platform. Two primary difficulties are: (1) Performance of interactive rendering, which is most computationally intensive portion of the visualization process. For terascale platforms, commodity clusters with graphics processors(GPUs) have been used for interactive rendering. For petascale platforms, visualization and rendering may be able to run efficiently on the supercomputer platform itself. (2) I/O bandwidth, which limits how much information can be written to disk. If we simply analyze the sparse information that is saved to disk we miss the opportunity to analyze the rich information produced every timestep by the simulation. For the first issue, we are pursuing in-situ analysis, in which simulations are coupled directly with analysis libraries at runtime. This milestone will evaluate the visualization and rendering performance of current and next generation supercomputers in contrast to GPU-based visualization clusters, and evaluate the performance of common analysis libraries coupled with the simulation that analyze and write data to disk during a running simulation. This milestone will explore, evaluate and advance the maturity level of these technologies and their applicability to problems of interest to the ASC program. Scientific simulation on parallel supercomputers is traditionally performed in four sequential steps: meshing, partitioning, solver, and visualization. Not all of these components are necessarily run on the supercomputer. In particular, the meshing and visualization typically happen on smaller but more interactive computing resources. However, the previous decade has seen a growth in both the need and ability to perform scalable parallel analysis, and this gives motivation for coupling the solver and visualization.

99 GENERAL AND MISCELLANEOUS↗

Message Passing and Shared Address Space Parallelism on an SMP Cluster

Currently, message passing (MP) and shared address space (SAS) are the two leading parallel programming paradigms. MP has been standardized with MPI, and is the more common and mature approach; however, code development can be extremely difficult, especially for irregularly structured computations. SAS offers substantial ease of programming, but may suffer from performance limitations due to poor spatial locality and high protocol overhead. In this paper, we compare the performance of and the programming effort required for six applications under both programming models on a 32-processor PC-SMP cluster, a platform that is becoming increasingly attractive for high-end scientific computing. Our application suite consists of codes that typically do not exhibit scalable performance under shared-memory programming due to their high communication-to-computation ratios and/or complex communication patterns. Results indicate that SAS can achieve about half the parallel efficiency of MPI for most of our applications, while being competitive for the others. A hybrid MPI+SAS strategy shows only a small performance advantage over pure MPI in some cases. Finally, improved implementations of two MPI collective operations on PC-SMP clusters are presented.

Shan, Hongzhang↗

Design and Implementation of Broadcast Algorithms for Extreme-Scale Systems

The scalability and performance of collective communication operations limit the scalability and performance of many scientific applications. This paper presents two new blocking and nonblocking Broadcast algorithms for communicators with arbitrary communication topology, and studies their performance. These algorithms benefit from increased concurrency and a reduced memory footprint, making them suitable for use on large-scale systems. Measuring small, medium, and large data Broadcasts on a Cray-XT5, using 24,576 MPI processes, the Cheetah algorithms outperform the native MPI on that system by 51%, 69%, and 9%, respectively, at the same process count. These results demonstrate an algorithmic approach to the implementation of the important class of collective communications, which is high performing, scalable, and also uses resources in a scalable manner.

99 GENERAL AND MISCELLANEOUS↗

Automated Design Space Exploration with Aspen

Architects and applications scientists often use performance models to explore a multidimensional design space of architectural characteristics, algorithm designs, and application parameters. With traditional performance modeling tools, these explorations forced users to first develop a performance model and then repeatedly evaluate and analyze the model manually. These manual investigations proved laborious and error prone. More importantly, the complexity of this traditional process often forced users to simplify their investigations. To address this challenge of design space exploration, we extend our Aspen (Abstract Scalable Performance Engineering Notation) language with three new language constructs: user-defined resources, parameter ranges, and a collection of costs in the abstract machine model. Then, we use these constructs to enable automated design space exploration via a nonlinear optimization solver. We show how four interesting classes of design space exploration scenarios can be derived from Aspen models and formulated as pure nonlinear programs. The analysis tools are demonstrated using examples based on Aspen models for a three-dimensional Fast Fourier Transform, the CoMD molecular dynamics proxy application, and the DARPA Streaming Sensor Challenge Problem. Our results show that this approach can compose and solve arbitrary performance modeling questions quickly and rigorously when compared to the traditional manual approach.

Spafford, Kyle L.↗

Enabling Graph Mining in RDF Triplestores using SPARQL for Holistic In-situ Graph Analysis

The graph analysis is now considered as a promising technique to discover useful knowledge in data with a new perspective. We envi- sion that there are two dimensions of graph analysis: OnLine Graph Analytic Processing (OLGAP) and Graph Mining (GM) where each respectively focuses on subgraph pattern matching and automatic knowledge discovery in graph. Moreover, as these two dimensions aim to complementarily solve complex problems, holistic in-situ graph analysis which covers both OLGAP and GM in a single system is critical for minimizing the burdens of operating multiple graph systems and transferring intermediate result-sets between those systems. Nevertheless, most existing graph analysis systems are only capable of one dimension of graph analysis. In this work, we take an approach to enabling GM capabilities (e.g., PageRank, connected-component analysis, node eccentricity, etc.) in RDF triplestores, which are originally developed to store RDF datasets and provide OLGAP capability. More specifically, to achieve our goal, we implemented six representative graph mining algorithms using SPARQL. The approach allows a wide range of available RDF data sets directly applicable for holistic graph analysis within a system. For validation of our approach, we evaluate performance of our implementations with nine real-world datasets and three different computing environments - a laptop computer, an Amazon EC2 instance, and a shared-memory Cray XMT2 URIKA-GD graph-processing appliance. The experimen- tal results show that our implementation can provide promising and scalable performance for real world graph analysis in all tested environments. The developed software is publicly available in an open-source project that we initiated.

97 MATHEMATICS AND COMPUTING↗

Microwave Irradiation Intensified Process for Scalable Functional Device Assembly

The environment pollution is a serious issue worldwide for human beings, which calls for ever-increasing demand towards environmental protection. Efficient enabling and utilization of various functional devices represent an essential solution to pursue the environment sustainability. Nanotechnology is emerging as an effective approach to improve the efficiency of functional devices for a variety of environmental applications. For example, the nanomaterials-based catalysts are being widely used in air pollution control, enabling catalysts with significantly enhanced performance. Scalable integration of functional nanostructural entities such as nanoparticles, nanowires, nanosheets into applicable three-dimensional (3-D) platform holds the keys of industrial-relevant manufacture of high-efficiency catalytic devices for emission control. Among all the integration approaches, the wet-chemical based integration is one of the top options due to its advantages in low-cost, flexibility in substrate geometry and composition, ease of operation, and high repeatability. The wet-chemical based processes usually require continuous heating to initiate the chemical reactions, which is regularly achieved by conventional resistive heating. However, such conventional heating method is plaqued by a poor heating efficiency. Moreover, the conventional heating is not suitable for those substrates with three-dimensional (3D) or complex geometry since it hardly provides sufficient heat transfer and mass transport within substrates, leading to poor uniformity of nanostructure deposition and low production rate. Thus, there is urgent demand to develop high-efficiency wet-chemical based integration strategy for nanostructured functional devices. In this project, a novel microwave-irradiation-intensified hydrothermal manufacturing process was developed for scalable manufacture of nanostructured monolithic catalytic devices. A wide array of metal oxide nanostructured coatings has been in situ grown on various 3D substrates for emission control applicaitons such as cordierite honeycombs and silicon carbide honeycomb. The growth efficiency, production rate, and distribution uniformity of nanostructure coatings obtained from microwave heating and conventional heating are investigated, which reveals the boosting effect of microwave irradiation on hydrothermal reactions. The silicon carbide honeycomb is found to be more favorable to microwave adsorption in contrast to cordierite honeycomb, and the corresponding reaction mechanism is studied. The obtained catalysts are used for catalytic oxidation of Volatile Organic Compounds (VOCs), showing excellent catalytic performance at low temperature for various compounds. As the proof of concept, the design of integrated manufacturing system for the microwave irradiation assisted continuous production of nanostructure array based monolithic catalysts is presented, which holds the key to mass production of monolithic catalysts in near future.

42 ENGINEERING↗

On Parallelizing Single Dynamic Simulation Using HPC Techniques and APIs of Commercial Software

Time-domain simulations are heavily used in today’s planning and operation practices to assess power system transient stability and post-transient voltage/frequency profiles following severe contingencies to comply with industry standards. Because of the increased modeling complexity, it is several times slower than real time for state-of-the-art commercial packages to complete a dynamic simulation for a large-scale model. With the growing stochastic behavior introduced by emerging technologies, power industry has seen a growing need for performing security assessment in real time. This paper presents a parallel implementation framework to speed up a single dynamic simulation by leveraging the existing stability model library in commercial tools through their application programming interfaces (APIs). Several high performance computing (HPC) techniques are explored such as parallelizing the calculation of generator current injection, identifying fast linear solvers for network solution, and parallelizing data outputs when interacting with APIs in the commercial package, TSAT. The proposed method has been tested on a WECC planning base case with detailed synchronous generator models and exhibits outstanding scalable performance with sufficient accuracy.

Diao, Ruisheng↗

Application of Cray XMT for Power Grid Contingency Selection

Contingency analysis is a key function to assess the impact of various combinations of power system component failures. It involves combinatorial numbers of contingencies which exceed the capability of even very large supercomputing platforms. Therefore, it is critical to select contingency cases within the constraint of computing power. This paper presents a contingency selection method employing graph theory (edge betweenness centrality) to power grid weighted graphs to remove low-impact components. The parallel implementation of the method was successfully carried out on the Cray XMT machine. The implementation takes advantage of the superior capabilities of the Cray XMT for graph-based problems and its programming features. This paper presents the performance scalability of Cray XMT and comparison with a cache-based, shared-memory machine.

24 POWER TRANSMISSION AND DISTRIBUTION↗

hPIC2: A hardware-accelerated, hybrid particle-in-cell code for dynamic plasma-material interactions

The exascale era of high performance computing promises to bring the field of computational plasma physics ever closer to the goal of accurate multiscale modeling. Such computers will rely on hardware acceleration to offload work to dedicated components, notably general-purpose graphics processing units (GPUs). However, devices from different manufacturers require software to be written with different parallel programming models, greatly increasing the code maintenance burden of applications designed to perform on more than one such device. hPIC2 is a hybrid plasma simulation code developed with the Kokkos performance portability framework to target the architectures that will drive exascale computing for the foreseeable future. As a hybrid simulation code, hPIC2 investigates the simultaneous use of various plasma models on the same domain, at the same time. hPIC2 also optionally couples to RustBCA, which accurately models ion-material interactions using the binary collision approximation (BCA) method. In conclusion, hPIC2 therefore achieves scalable performance on a variety of computing architectures when simulating complex and diverse plasmas, particularly near plasma-material interfaces.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Efficient I/O for Neural Network Training with Compressed Data

FanStore is a shared object store that enables efficient and scalable neural network training on supercomputers. By providing a global cache layer on node-local burst buffers using a compressed representation, it significantly enhances the processing capability of deep learning (DL) applications on existing hardware. In addition, FanStore allows POSIX-compliant file access to the compressed data in user space. We investigate the tradeoff between runtime overhead and data compression ratio using real-world datasets and applications, and propose a compressor selection algorithm to maximize storage capacity given performance constraints. We consider both asynchronous (i.e., with prefetching) and synchronous I/O strategies, and propose mechanisms for selecting compressors for both approaches. Using FanStore, the same storage hardware can host 2-13x more data for example applications without significant runtime overhead. Empirically, our experiments show that FanStore scales to 512 compute nodes with near linear performance scalability.

Zhang, Zhao↗

Summer Internship Report: ARA2 Benchmarking

Over the past decade, the RISC-V Instruction Set Architecture (ISA) has emerged as a significant player in both academic and industrial processor design due to its open-source nature, modular extension system, and versatility across domains ranging from microcontrollers to high-performance computing (HPC). One of its most important recent advancements is the RISC-V Vector Extension (RVV), which enables explicit data-level parallelism through vector registers and vectorized instructions. Unlike traditional SIMD (Single Instruction, Multiple Data) architectures that fix vector lengths at design time, RVV uses the concept of VLEN (vector register length) as a hardware-independent parameter and allows software to adapt dynamically to the available vector width. This flexible approach ensures portability across implementations while enabling scalable performance. The ARA2 core is a parameterizable RISC-V vector processor developed at the Integrated Systems Lab at ETH Zürich and the University of Bologna. Designed as a tightly-coupled accelerator to a scalar RISC-V core, ARA2 implements the RVV 1.0 specification and offers tunable architectural parameters such as the number of vector lanes, VLEN, and cache sizes.

97 MATHEMATICS AND COMPUTING↗

Robotic Specialization in Autonomous Robotic Structural Assembly

Robotic in-space assembly of large space structures is a long-term NASA goal to reduce launch costs and enable larger scale missions. Recently, researchers have proposed using discrete lattice building blocks and co-designed robots to build high-performance, scalable primary structure for various on-orbit and surface applications. These robots would locomote on the lattice and work in teams to build and reconfigure building-blocks into functional structure. However, the most reliable and efficient robotic system architecture, characterized by the number of different robotic 'species' and the allocation of functionality between species, is an open question. To address this problem, we decompose the robotic building-block assembly task into functional primitives and, in simulation, study the performance of the the variety of possible resulting architectures. For a set consisting of five process types (move self, move block, move friend, align bock, fasten block), we describe a method of feature space exploration and ranking based on energy and reliability cost functions. The solution space is enumerated, filtered for unique solutions, and evaluated against energy and reliability cost functions for various simulated build sizes. We find that a 2 species system, dividing the five mentioned process types between one unit cell transport robot and one fastening robot, results in the lowest energy cost system, at some cost to reliability. This system enables fastening functionality to occupy the build front while reducing the need for that functional mass to travel back and forth from a feed station. Because the details of a robot design affect the weighting and final allocation of functionality, a sensitivity analysis was conducted to evaluate the effect of changing mass allocations on architecture performance. Future systems with additional functionalities such as repair, inspection, and others may use this process to analyze and determine alternative robot architectures.

Bernus, Borbala↗