Search NASA⌕ Search

SEARCH · Search NASA

Results for “Computational 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 1,657 records · Page 92

Some Problems and Solutions in Transferring Ecosystem Simulation Codes to Supercomputers

Many computer codes for the simulation of ecological systems have been developed in the last twenty-five years. This development took place initially on main-frame computers, then mini-computers, and more recently, on micro-computers and workstations. Recent recognition of ecosystem science as a High Performance Computing and Communications Program Grand Challenge area emphasizes supercomputers (both parallel and distributed systems) as the next set of tools for ecological simulation. Transferring ecosystem simulation codes to such systems is not a matter of simply compiling and executing existing code on the supercomputer since there are significant differences in the system architectures of sequential, scalar computers and parallel and/or vector supercomputers. To more appropriately match the application to the architecture (necessary to achieve reasonable performance), the parallelism (if it exists) of the original application must be exploited. We discuss our work in transferring a general grassland simulation model (developed on a VAX in the FORTRAN computer programming language) to a Cray Y-MP. We show the Cray shared-memory vector-architecture, and discuss our rationale for selecting the Cray. We describe porting the model to the Cray and executing and verifying a baseline version, and we discuss the changes we made to exploit the parallelism in the application and to improve code execution. As a result, the Cray executed the model 30 times faster than the VAX 11/785 and 10 times faster than a Sun 4 workstation. We achieved an additional speed-up of approximately 30 percent over the original Cray run by using the compiler's vectorizing capabilities and the machine's ability to put subroutines and functions "in-line" in the code. With the modifications, the code still runs at only about 5% of the Cray's peak speed because it makes ineffective use of the vector processing capabilities of the Cray. We conclude with a discussion and future plans.

Skiles, J. W.↗

Energy-efficient scientific computing using chemical reservoirs

The rapid growth of computing demands driven by scientific computing, data analytics, and artificial intelligence (AI) advancements has exposed the limitations of traditional digital processing systems. These systems are nearing physical energy barriers, making significant gains in energy efficiency increasingly unattainable. As we advance toward post-exascale computing, disruptive approaches are critical to overcoming these limitations. Among emerging analog solutions, biochemical computing offers a transformative path for achieving orders-of-magnitude improvements in energy efficiency. By leveraging the natural optimization capabilities of chemical reaction networks (CRNs), biochemical systems have the potential to meet high-performance computing needs through natural scalability. However, numerous challenges remain, including theoretical limitations in mapping computational problems to CRNs and practical barriers in implementing biochemical computing devices. In this paper, we present a framework for chemical computation using biochemical systems and introduce key components of our approach for energy-efficient scientific computing. We showcase the feasibility of this framework by solving a system of ordinary differential equations by emulating a chemical reservoir device, demonstrating its potential for addressing modern computing challenges. This work lays a foundational step toward harnessing the computational power of chemistry to design energy-efficient, scalable, high-performance next-generation computing systems.

Johnson, Connah G. M. [Pacific Northwest National ↗

ExaTN: Scalable GPU-Accelerated High-Performance Processing of General Tensor Networks at Exascale

We present ExaTN (Exascale Tensor Networks), a scalable GPU-accelerated C++ library which can express and process tensor networks on shared- as well as distributed-memory high-performance computing platforms, including those equipped with GPU accelerators. Specifically, ExaTN provides the ability to build, transform, and numerically evaluate tensor networks with arbitrary graph structures and complexity. It also provides algorithmic primitives for the optimization of tensor factors inside a given tensor network in order to find an extremum of a chosen tensor network functional, which is one of the key numerical procedures in quantum many-body theory and quantum-inspired machine learning. Numerical primitives exposed by ExaTN provide the foundation for composing rather complex tensor network algorithms. We enumerate multiple application domains which can benefit from the capabilities of our library, including condensed matter physics, quantum chemistry, quantum circuit simulations, as well as quantum and classical machine learning, for some of which we provide preliminary demonstrations and performance benchmarks just to emphasize a broad utility of our library.

97 MATHEMATICS AND COMPUTING↗

Extreme-scale workflows: A perspective from the JLESC international community

The Joint Laboratory for Extreme-Scale Computing (JLESC) focuses on software challenges in high-performance computing systems to meet the needs of today’s science campaigns, which often require large resources, consist of multiple tasks, and generate vast amounts of data. In this context, extreme-scale workflows have been the key factor in enabling scientific discoveries by helping scientists automate the dependencies and data exchanges between workflow tasks, instead of managing those manually. Here, in this paper, we present representative extreme-scale workflows and feature workflow systems developed by JLESC participating institutions. We present lessons learned while developing these tools, alongside with the open challenges and future research directions in the field of extreme-scale workflows.

97 MATHEMATICS AND COMPUTING↗

Density scaling approximation for Monte-Carlo simulations of radioactive plumes

The release of radioactive gas into the atmosphere can diffuse into large volumes of air downwind from the point of release. The extent of radioactivity can cover thousands of cubic meters of air. For such large volumes, the weather models used to predict the down-wind distribution of the plume and the radiation transport models used to predict the radiation reaching ground-level from the plume can take tens of hours of computer time on multi-node institutional High-Performance Computing facilities. In this paper we focus on the radiation transport aspect of plume modeling. Here, we describe a phenomenological method for approximating the amounts of radiation that reach ground level from large volumes of a static radioactive plume that can be calculated on a stand-alone personal computer in much shorter computation times than those usually needed for such large volume evaluations. We refer to this method as the Density Scaling Approximation (DSA). Its ability to approximate ground-level count rates of large plumes comes from using a small-plume volume with a scaled-up value of air density to simulate the same number of scatterings that occur during transport in larger plume volumes at normal air density. We demonstrate the DSA by using a 100 m-diameter air-filled hemispherical dome geometry with a uniform volumetric activity of 135 Xe gas throughout the air-filled volume. The DSA for a larger dome diameter is obtained by evaluating the 100 m dome with an air density scaled up by the linear ratio of the larger diameter to the 100 m diameter. We find that this approximation works well for dome diameters up to 1200 m – the largest diameter studied and a size more than sufficient for accounting for all the radiation from 135 Xe. Moreover, most of our DSA results can be calculated over 500 times faster than corresponding full-sized geometry with normal air density. To help evaluate the accuracy of the DSA and gain insight into how well it can reproduce different regions of the spectra, we use three, easily understood regions of interest to compare the DSA results to the full-sized geometry at normal air density results. These regions are the full-energy peak, the region of single-Compton scattering, and the region of multiple-Compton scattering. We show how the dominance of the Compton scattering mechanism determines this division and thus provides insight into how Compton scattering is manifested in spectra from photon scattering through air in general, and how well the DSA approximation works.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Experiments and Analyses of Data Transfers Over Wide-Area Dedicated Connections

Dedicated wide-area network connections are increasingly employed in high-performance computing and big data scenarios. One might expect the performance and dynamics of data transfers over such connections to be easy to analyze due to the lack of competing traffic. However, non-linear transport dynamics and end-system complexities (e.g., multi-core hosts and distributed filesystems) can in fact make analysis surprisingly challenging. We present extensive measurements of memory-to-memory and disk-to-disk file transfers over 10 Gbps physical and emulated connections with 0–366 ms round trip times (RTTs). For memory-to-memory transfers, profiles of both TCP and UDT throughput as a function of RTT show concave and convex regions; large buffer sizes and more parallel flows lead to wider concave regions, which are highly desirable. TCP and UDT both also display complex throughput dynamics, as indicated by their Poincare maps and Lyapunov exponents. For disk-to-disk transfers, we determine that high throughput can be achieved via a combination of parallel I/O threads, parallel network threads, and direct I/O mode. Our measurements also show that Lustre filesystems can be mounted over long-haul connections using LNet routers, although challenges remain in jointly optimizing file I/O and transport method parameters to achieve peak throughput.

Rao, Nageswara S.↗

TOWARDS RAPID RESPONSE UPDATES OF POPULATIONS AT RISK

Understanding population at risks has been a focus of the LandScan program through its development of population estimates. With advancements in computer vision, deep learning technologies and access to High Performance Computing (HPC) and high resolution imagery, population estimates are now modeled at the building level. However, when those patterns are disrupted, rapid updates to population distribution estimates are needed to support humanitarian aid and response. Oak Ridge National Laboratory (ORNL) recently adapted an existing deep learning building footprint extraction model in development of a scalable approach to Building Damage Assessments (BDA). This new opportunity opens the possibility of automating BDA to support rapid population distribution estimate updates for geographic areas involved in geopolitical conflicts or natural events for humanitarian aid and response or where to focus recovery efforts. In addition, incorporate social surveys to further model human behavior under conflict or other scenarios that disrupt normal patterns of life.

Urban, Marie↗

Evaluation of Counter-Based Dynamic Load Balancing Schemes for Massive Contingency Analysis on Over 10,000 Cores

Contingency analysis studies are necessary to assess the impact of possible power system component failures. The results of the contingency analysis are used to ensure the grid reliability, and in power market operation for the feasibility test of market solutions. Currently, these studies are performed in real time based on the current operating conditions of the grid with a set of pre-selected contingency list, which might result in overlooking some critical contingencies caused by variable system status. To have a complete picture of a power grid, more contingencies need to be studied to improve grid reliability. High-performance computing techniques hold the promise of being able to perform the analysis for more contingency cases within a much shorter time frame. This paper evaluates the performance of counter-based dynamic load balancing schemes for a massive contingency analysis program on 10,000+ cores. One million N-2 contingency analysis cases with a Western Electricity Coordinating Council power grid model have been used to demonstrate the performance. The speedup of 3964 with 4096 cores and 7877 with 10240 cores are obtained. This paper reports the performance of the load balancing scheme with a single counter and two counters, describes disk I/O issues, and discusses other potential techniques for further improving the performance.

Balancing Schemes for Massive Contingency↗

RLScheduler: An Automated HPC Batch Job Scheduler Using Reinforcement Learning

Today’s high-performance computing (HPC) platforms are still dominated by batch jobs. Accordingly, effective batch job scheduling is crucial to obtain high system efficiency. Existing HPC batch job schedulers typically leverage heuristic priority functions to prioritize and schedule jobs. But, once configured and deployed by the experts, such priority functions can hardly adapt to the changes of job loads, optimization goals, or system settings, potentially leading to degraded system efficiency when changes occur. To address this fundamental issue, we present RLScheduler, an automated HPC batch job scheduler built on reinforcement learning. RLScheduler relies on minimal manual interventions or expert knowledge, but can learn high-quality scheduling policies via its own continuous ‘trial and error’. We introduce a new kernel-based neural network structure and trajectory filtering mechanism in RLScheduler to improve and stabilize the learning process. Through extensive evaluations, we confirm that RLScheduler can learn high-quality scheduling policies towards various workloads and various optimization goals with relatively low computation cost. Moreover, we show that the learned models perform stably even when applied to unseen workloads, making them practical for production use.

Zhang, Di↗

Mojo Application Monitoring Tool Suite

Mojo is a software tool suite that can be used to monitor the progress of compute jobs on Linux Clusters and other high-performance computing platforms. Mojo is designed to allow system administrators to monitor the health and progress of computing jobs, and to allow users to view the progress and status of their own jobs. The facilities provided include the ability to notify users of job "hangs", and to take an automated action (e.g killing the job) when something goes wrong. These operations can lead to a more efficient use of scarce resources.

Ballance, Robert↗

Lightweight performance data collectors 2.0 with Eiger support.

We report on the use and design of a portable, extensible performance data collection tool motivated by modeling needs of the high performance computing systems co-design com- munity. The lightweight performance data collectors with Eiger support is intended to be a tailorable tool, not a shrink-wrapped library product, as pro ling needs vary widely. A single code markup scheme is reported which, based on compilation ags, can send perfor- mance data from parallel applications to CSV les, to an Eiger mysql database, or (in a non-database environment) to at les for later merging and loading on a host with mysql available. The tool supports C, C++, and Fortran applications.

Allan, Benjamin A.↗

Phase I Final Technical Report on Energy-Efficient Reconfigurable Universal Accelerator Interconnect

It is well known that application specific computing systems, optimally designed and configured for a given workload, offer much higher energy-efficiency and throughput than general purpose systems. In modern computing systems, heterogeneous computing systems have emerged that exploit the energy and performance benefits of combining various different domain-specific processor architectures. Application domains such as high-performance computing and machine learning now process terabyte-sized data sets, requiring enormous processing and memory resources. These applications have very high-power consumption due to bottlenecks in the electrical interconnection between processing units. This project aims to reduce both communication energy and latency by integrating universally available accelerators with silicon photonics. It also aims to increase system throughput by exploiting emerging technologies in silicon photonic reconfigurable interconnects; this will allow the system to balance itself in real time to accommodate changes in workloads and data flows.

2.5D/3D integration↗

AI-Driven Accelerated Inclusion Analysis for Energy Efficient Steelmaking (Final CRADA Report)

This was a collaborative effort between Lawrence Livermore National Security, LLC (LLNS) as manager and operator of Lawrence Livermore National Laboratory (LLNL) and ArcelorMittal USA Research LLC (“ArcelorMittal” as the Participant), to use scanning electron microscopy (SEM) images, computer vision and machine learning methods, and high-performance computing to accelerate the inclusion analysis process of liquid steel so that new methods can be used for near-real time process control on the shop floor.

36 MATERIALS SCIENCE↗

Properties of Electronic Materials

This final technical report summarizes the research conducted under DOE Grant DE-SC0002623, "Properties of Electronic Materials," led by Principal Investigator Shengbai Zhang at Rensselaer Polytechnic Institute. Over the 16-year period, the project employed first-principles computational methods to investigate the structural, electronic, and dynamic properties of a wide range of electronic materials, with applications in energy technologies, optoelectronics, and data storage. Key areas included topological insulators, phase-change materials, graphene and two-dimensional systems, perovskites for photovoltaics, defect engineering in semiconductors, kagome lattices, and ultrafast carrier dynamics. The research resulted in 115 peer-reviewed publications, advancing fundamental understanding of material behaviors at the atomic scale and contributing to innovations in renewable energy, memory devices, and quantum materials. Findings have implications for improving energy efficiency, developing lead-free solar cells, and enabling high-speed data processing. The work has trained numerous graduate students and postdocs, fostering the next generation of computational materials scientists. The original goals were to develop theoretical models and computational tools to predict and optimize electronic properties of materials for energy applications. All objectives were accomplished, with no major departures from planned methodologies. Challenges in computational scaling were addressed through access to high-performance computing resources.

36 MATERIALS SCIENCE↗

NASKERN Program Tests Computers

Computer program, NASKERN developed to aid in evaluation of performances of supercomputers. Consists of seven test-kernel programs performing computations typical of supercomputer calculations. Written in FORTRAN 77.

Bailey, David H.↗

Global Magnetohydrodynamic Modeling of the Solar Corona

Under this contract, we have continued our investigations of the large scale structure of the solar corona and inner heliosphere using global magnetohydrodynamic (MHD) simulations. These computations have also formed the basis for studies of coronal mass ejections (CMES) using realistic coronal configurations. We have developed a technique for computing realistic magnetohydrodynamic (MHD) computations of the solar corona and inner heliosphere. To perform computations that can be compared with specific observations, it is necessary to incorporate solar observations into the boundary conditions. We have used the Wilcox Solar Observatory synoptic maps (collected during a solar rotation by daily measurements of the line-of-sight magnetic field at central meridian) to specify the radial magnetic field (B,) at the photosphere. For the initial condition, we use a potential magnetic field consistent with the specified distribution of B, at the lower boundary, and a wind solution consistent with the specified plasma density and temperature at the solar surface. Together this initial condition forms a (non-equilibrium) approximation of the state of the solar corona for the time-dependent MHD computation. The MHD equations are then integrated in time to steady state. Here we describe solutions relevant to a recent solar eclipse, as well as Ulysses observations. We have also developed a model configuration of solar minimum, useful for studying CME initiation and propagation.

Linker, Jon A.↗

A Geometry Based Infra-Structure for Computational Analysis and Design

The computational steps traditionally taken for most engineering analysis suites (computational fluid dynamics (CFD), structural analysis, heat transfer and etc.) are: (1) Surface Generation -- usually by employing a Computer Assisted Design (CAD) system; (2) Grid Generation -- preparing the volume for the simulation; (3) Flow Solver -- producing the results at the specified operational point; (4) Post-processing Visualization -- interactively attempting to understand the results. For structural analysis, integrated systems can be obtained from a number of commercial vendors. These vendors couple directly to a number of CAD systems and are executed from within the CAD Graphical User Interface (GUI). It should be noted that the structural analysis problem is more tractable than CFD; there are fewer mesh topologies used and the grids are not as fine (this problem space does not have the length scaling issues of fluids). For CFD, these steps have worked well in the past for simple steady-state simulations at the expense of much user interaction. The data was transmitted between phases via files. In most cases, the output from a CAD system could go to Initial Graphics Exchange Specification (IGES) or Standard Exchange Program (STEP) files. The output from Grid Generators and Solvers do not really have standards though there are a couple of file formats that can be used for a subset of the gridding (i.e. PLOT3D data formats). The user would have to patch up the data or translate from one format to another to move to the next step. Sometimes this could take days. Specifically the problems with this procedure are:(1) File based -- Information flows from one step to the next via data files with formats specified for that procedure. File standards, when they exist, are wholly inadequate. For example, geometry from CAD systems (transmitted via IGES files) is defined as disjoint surfaces and curves (as well as masses of other information of no interest for the Grid Generator). This is particularly onerous for modern CAD systems based on solid modeling. The part was a proper solid and in the translation to IGES has lost this important characteristic. STEP is another standard for CAD data that exists and supports the concept of a solid. The problem with STEP is that a solid modeling geometry kernel is required to query and manipulate the data within this type of file. (2) 'Good' Geometry. A bottleneck in getting results from a solver is the construction of proper geometry to be fed to the grid generator. With 'good' geometry a grid can be constructed in tens of minutes (even with a complex configuration) using unstructured techniques. Adroit multi-block methods are not far behind. This means that a million node steady-state solution can be computed on the order of hours (using current high performance computers) starting from this 'good' geometry. Unfortunately, the geometry usually transmitted from the CAD system is not 'good' in the grid generator sense. The grid generator needs smooth closed solid geometry. It can take a week (or more) of interaction with the CAD output (sometimes by hand) before the process can begin. One way Communication. (3) One-way Communication -- All information travels on from one phase to the next. This makes procedures like node adaptation difficult when attempting to add or move nodes that sit on bounding surfaces (when the actual surface data has been lost after the grid generation phase). Until this process can be automated, more complex problems such as multi-disciplinary analysis or using the above procedure for design becomes prohibitive. There is also no way to easily deal with this system in a modular manner. One can only replace the grid generator, for example, if the software reads and writes the same files. Instead of the serial approach to analysis as described above, CAPRI takes a geometry centric approach. This makes the actual geometry (not a discretized version) accessible to all phases of the analysis. The connection to the geometry is made through an Application Programming Interface (API) and NOT a file system. This API isolates the top-level applications (grid generators, solvers and visualization components) from the geometry engine. Also this allows the replacement of one geometry kernel with another, without effecting these top-level applications. For example, if UniGraphics is used as the CAD package then Parasolid (UG's own geometry engine) can be used for all geometric queries so that no solid geometry information is lost in a translation. This is much better than STEP because when the data is queried, the same software is executed as used in the CAD system. Therefore, one analyzes the exact part that is in the CAD system. CAPRI uses the same idea as the commercial structural analysis codes but does not specify control. Software components of the CAD system are used, but the analysis suite, not the CAD operator, specifies the control of the software session. This also means that the license issues (may be) minimized and individuals need not have to know how to operate a CAD system in order to run the suite.

Haimes, Robert↗

Post-Optimality Analysis In Aerospace Vehicle Design

This analysis pertains to the applicability of optimal sensitivity information to aerospace vehicle design. An optimal sensitivity (or post-optimality) analysis refers to computations performed once the initial optimization problem is solved. These computations may be used to characterize the design space about the present solution and infer changes in this solution as a result of constraint or parameter variations, without reoptimizing the entire system. The present analysis demonstrates that post-optimality information generated through first-order computations can be used to accurately predict the effect of constraint and parameter perturbations on the optimal solution. This assessment is based on the solution of an aircraft design problem in which the post-optimality estimates are shown to be within a few percent of the true solution over the practical range of constraint and parameter variations. Through solution of a reusable, single-stage-to-orbit, launch vehicle design problem, this optimal sensitivity information is also shown to improve the efficiency of the design process, For a hierarchically decomposed problem, this computational efficiency is realized by estimating the main-problem objective gradient through optimal sep&ivity calculations, By reducing the need for finite differentiation of a re-optimized subproblem, a significant decrease in the number of objective function evaluations required to reach the optimal solution is obtained.

Braun, Robert D.↗