Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “computer hardware”

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 1,369 records · Page 76

Mastering HPC Runtime Prediction: From Observing Patterns to a Methodological Approach

The continual expansion of high-performance computing (HPC) brings with it an increasing need for efficiency. Heavy investment in energy, hardware, and software infrastructure to support peta- and exascale computing requires the optimization of existing systems and, wherever possible, the discernment and adoption of best-practices towards these goals. Such is the case for runtime prediction. When a job is submitted to an HPC system, an estimate of its runtime is provided by the user in the form of "requested wallclock". Error in this user-provided estimate can lead to jobs being prematurely killed by the scheduler, increased wait time on the queue, and decreased system utilization. More than fifteen years of research has been directed at mitigating these effects by using data-driven runtime predictions. Codified here is a set of commonalities and insights emerging from this body of work, which we present as recommendations and best practices. These practices are combined into a methodological approach described and evaluated on an 11-million-job dataset from the National Renewable Energy Laboratory's petascale HPC system, Eagle. This dataset and the accompanying codebase have been released to the public domain for the benefit of the wider HPC research community.

high performance computing↗

Performance testing of the Galileo attitude control system

Performance testing plans, procedures, and initial results for the Galileo Attitude and Articulation Control Subsystem are described. The objectives of performance testing include test bed comparisons, examining the interactions between spacecraft dynamics and controllers, and verifying the ability to tune control algorithms in flight to meet specific performance requirements. Two primary test beds are described. One, the Integration Test Laboratory, drives actual flight hardware components and support equipment with a real time computer simulation of spacecraft dynamics. The other, the Functional Simulator, is a nonreal time computer simulation with high fidelity spacecraft dynamics, including flexible appendages, fuel slosh, and simulated hardware components. Initial Functional Simulator results are presented which illustrate spacecraft response to open loop spin-up and turn commands. A closed loop sun acquisition turn is also demonstrated. Future tests for verifying performance of attitude determination, spacecraft control, and scan platform pointing functions are outlined.

Bell, C. E.↗

Mastering HPC Runtime Prediction: From Observing Patterns to a Methodological Approach: Preprint

The continual expansion of high-performance computing (HPC) brings with it an increasing need for efficiency. Heavy investment in energy, hardware, and software infrastructure to support peta- and exascale computing requires the optimization of existing systems and, wherever possible, the discernment and adoption of best-practices towards these goals. Such is the case for runtime prediction. When a job is submitted to an HPC system, an estimate of its runtime is provided by the user in the form of "requested wallclock''. Error in this user-provided estimate can lead to jobs being prematurely killed by the scheduler, increased wait time on the queue, and decreased system utilization. More than fifteen years of research has been directed at mitigating these effects by using data-driven runtime predictions. Codified here is a set of commonalities and insights emerging from this body of work, which we present as recommendations and best practices. These practices are combined into a methodological approach described and evaluated on an 11-million-job dataset from the National Renewable Energy Laboratory's petascale HPC system, Eagle. This dataset and the accompanying codebase have been released to the public domain for the benefit of the wider HPC research community.

high performance computing↗

Empowering Scientific Innovation Through An Integrated Research Infrastructure: The Role of the Advanced Computing Ecosystem

As the landscape of computational science evolves, the Department of Energy (DOE) is reimagining the roles of its large-scale computing facilities to meet emerging research challenges. The Integrated Research Infrastructure (IRI) program aims to transform how experiments are designed, conducted, and shared, with significant impacts on all stakeholders. In response, the Oak Ridge Leadership Computing Facility (OLCF) has established the Advanced Computing Ecosystem (ACE), a strategic framework to prepare its hardware, software, and experimental capabilities for the IRI era. ACE focuses on integrating novel compute environments, orchestrating advanced workflows, and developing foundational technologies, ensuring a seamless transition to IRI while accelerating scientific discovery. This paper outlines ACE's role in advancing OLCF's mission and its impact on the future of computational science.

Widener, Patrick↗

Self-consistent mean-field quantum approximate optimization

We introduce a self-consistent mean-field quantum optimization algorithm that approximates the ground state of classical Ising Hamiltonians. The algorithm decomposes the problem into independent subproblems and treats the interactions between them in a mean-field manner. These interactions are captured by a common environment, constructed self-consistently through a variational quantum circuit, and which modifies the subproblems to account for mutual influence while maintaining computational independence. Consequently, subproblems can be solved individually, avoiding the computational cost of the full problem. We explore the properties of the generated environment and assess the algorithm's performance through extensive numerical simulations on Sherrington-Kirkpatrick spin glasses. Furthermore, we apply it experimentally to a weighted maximum clique problem applied to molecular docking. This framework enables the solution of problems that would otherwise exceed the qubit and gate counts of current quantum hardware.

Dupont, Maxime [Rigetti Computing] (ORCID:00000001↗

Self-consistent mean-field quantum approximate optimization

We introduce a self-consistent mean-field quantum optimization algorithm that approximates the ground state of classical Ising Hamiltonians. The algorithm decomposes the problem into independent subproblems and treats the interactions between them in a mean-field manner. These interactions are captured by a common environment, constructed self-consistently through a variational quantum circuit, and which modifies the subproblems to account for mutual influence while maintaining computational independence. Consequently, subproblems can be solved individually, avoiding the computational cost of the full problem. We explore the properties of the generated environment and assess the algorithm's performance through extensive numerical simulations on Sherrington-Kirkpatrick spin glasses. Furthermore, we apply it experimentally to a weighted maximum clique problem applied to molecular docking. This framework enables the solution of problems that would otherwise exceed the qubit and gate counts of current quantum hardware.

Dupont, Maxime [Rigetti Computing] (ORCID:00000001↗

Self-consistent mean-field quantum approximate optimization

We introduce a self-consistent mean-field quantum optimization algorithm that approximates the ground state of classical Ising Hamiltonians. The algorithm decomposes the problem into independent subproblems and treats the interactions between them in a mean-field manner. These interactions are captured by a common environment, constructed self-consistently through a variational quantum circuit, and which modifies the subproblems to account for mutual influence while maintaining computational independence. Consequently, subproblems can be solved individually, avoiding the computational cost of the full problem. We explore the properties of the generated environment and assess the algorithm's performance through extensive numerical simulations on Sherrington-Kirkpatrick spin glasses. Furthermore, we apply it experimentally to a weighted maximum clique problem applied to molecular docking. This framework enables the solution of problems that would otherwise exceed the qubit and gate counts of current quantum hardware.

Dupont, Maxime [Rigetti Computing] (ORCID:00000001↗

Toward coherent quantum computation of scattering amplitudes with a measurement-based photonic quantum processor

In recent years, applications of quantum simulation have been developed to study the properties of strongly interacting theories. This has been driven by two factors: on the one hand, needs from theorists to have access to physical observables that are prohibitively difficult to study using classical computing; on the other hand, quantum hardware becoming increasingly reliable and scalable to larger systems. In this work, we discuss the feasibility of using quantum optical simulation for studying scattering observables that are presently inaccessible via lattice QCD and are at the core of the experimental program at Jefferson Laboratory, the future Electron-Ion Collider, and other accelerator facilities. We show that recent progress in measurement-based photonic quantum computing can be leveraged to provide deterministic generation of required exotic gates and implementation in a single photonic quantum processor. Published by the American Physical Society 2024

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Computation Directorate 2008 Annual Report

Whether a computer is simulating the aging and performance of a nuclear weapon, the folding of a protein, or the probability of rainfall over a particular mountain range, the necessary calculations can be enormous. Our computers help researchers answer these and other complex problems, and each new generation of system hardware and software widens the realm of possibilities. Building on Livermore's historical excellence and leadership in high-performance computing, Computation added more than 331 trillion floating-point operations per second (teraFLOPS) of power to LLNL's computer room floors in 2008. In addition, Livermore's next big supercomputer, Sequoia, advanced ever closer to its 2011-2012 delivery date, as architecture plans and the procurement contract were finalized. Hyperion, an advanced technology cluster test bed that teams Livermore with 10 industry leaders, made a big splash when it was announced during Michael Dell's keynote speech at the 2008 Supercomputing Conference. The Wall Street Journal touted Hyperion as a 'bright spot amid turmoil' in the computer industry. Computation continues to measure and improve the costs of operating LLNL's high-performance computing systems by moving hardware support in-house, by measuring causes of outages to apply resources asymmetrically, and by automating most of the account and access authorization and management processes. These improvements enable more dollars to go toward fielding the best supercomputers for science, while operating them at less cost and greater responsiveness to the customers.

99 GENERAL AND MISCELLANEOUS↗

Development of a 32-bit UNIX-based ELAS workstation

A mini/microcomputer UNIX-based image analysis workstation has been designed and is being implemented to use the Earth Resources Laboratory Applications Software (ELAS). The hardware system includes a MASSCOMP 5600 computer, which is a 32-bit UNIX-based system (compatible with AT&T System V and Berkeley 4.2 BSD operating system), a floating point accelerator, a 474-megabyte fixed disk, a tri-density magnetic tape drive, and an 1152 by 910 by 12-plane color graphics/image interface. The software conversion includes reconfiguring the ELAs driver Master Task, recompiling and then testing the converted application modules. This hardware and software configuration is a self-sufficient image analysis workstation which can be used as a stand-alone system, or networked with other compatible workstations.

Spiering, Bruce A.↗

Multiple-body simulation with emphasis on integrated Space Shuttle vehicle

The program to obtain intergrid communications - Pegasus - was enhanced to make better use of computing resources. Periodic block tridiagonal and penta-diagonal diagonal routines in OVERFLOW were modified to use a better algorithm to speed up the calculation for grids with periodic boundary conditions. Several programs were added to collar grid tools and a user friendly shell script was developed to help users generate collar grids. User interface for HYPGEN was modified to cope with the changes in HYPGEN. ET/SRB attach hardware grids were added to the computational model for the space shuttle and is currently incorporated into the refined shuttle model jointly developed at Johnson Space Center and Ames Research Center. Flow simulation for the integrated space shuttle vehicle at flight Reynolds number was carried out and compared with flight data as well as the earlier simulation for wind tunnel Reynolds number.

Chiu, Ing-Tsau↗

Open Source Platform for Live Processing of High Data Rate Electron Microscopy (SBIR Phase I Award Topic 15, Subtopic B, Final Scientific / Technical Report)

4D scanning transmission electron microscopy (STEM) is experiencing a revolution both in terms of data size, and data veracity, as are many other areas of experimental data acquisition. This presents enormous opportunities to make new scientific discoveries, but is also hamstrung by aging software that is no longer fit for purpose. Experimental data acquisition platforms must make the same investments in software infrastructure that the simulation community made decades ago to take advantage of high-performance computing (HPC) if they hope to realize the full potential of the latest advances made in detector technology. This requires rethinking the design and deployment of hardware, moving away from small, fixed commodity computing capabilities with closed interfaces to open, modular, scalable systems designed for parallelism and throughput. The technical approach will leverage best-in-class open source technologies, and develop new approaches strategically to build an open source platform for experimental data acquisition targeting high data rate electron microscopy. The project will develop a software platform that uses the power and performance of modern HPC resources to ingest the large volume of data produced by electron microscopes using direct electron detectors. A set of configurable pipelines will be used to enable the processing to be tailored to the problem being investigated with builtin visualization of data at interactive rates. This project will leverage investments made by the DOE and other agencies to provide a powerful software platform for the experimental data community working with most powerful electron microscopes, such as 4D STEM detectors. It will create a platform that serves as a reference for the community. Permissive open source licensing will make the platform available to all, with a simple deployment strategy offering the ability to keep data close to where it is generated. This advanced software platform will require customization for each deployment, creating opportunities for software services that are typically offered to extend open source software platforms.

97 MATHEMATICS AND COMPUTING↗

XploreNAS : Explore Adversarially Robust and Hardware-efficient Neural Architectures for Non-ideal Xbars

Compute In-Memory platforms such as memristive crossbars are gaining focus as they facilitate acceleration of Deep Neural Networks (DNNs) with high area and compute efficiencies. However, the intrinsic non-idealities associated with the analog nature of computing in crossbars limits the performance of the deployed DNNs. Furthermore, DNNs are shown to be vulnerable to adversarial attacks leading to severe security threats in their large-scale deployment. Thus, finding adversarially robust DNN architectures for non-ideal crossbars is critical to the safe and secure deployment of DNNs on the edge. This work proposes a two-phase algorithm-hardware co-optimization approach called XploreNAS that searches for hardware efficient and adversarially robust neural architectures for non-ideal crossbar platforms. We use the one-shot Neural Architecture Search approach to train a large Supernet with crossbar-awareness and sample adversarially robust Subnets therefrom, maintaining competitive hardware efficiency. Our experiments on crossbars with benchmark datasets (SVHN, CIFAR10, CIFAR100) show up to ~8–16% improvement in the adversarial robustness of the searched Subnets against a baseline ResNet-18 model subjected to crossbar-aware adversarial training. We benchmark our robust Subnets for Energy-Delay-Area-Products (EDAPs) using the Neurosim tool and find that with additional hardware efficiency–driven optimizations, the Subnets attain ~1.5–1.6× lower EDAPs than ResNet-18 baseline.

97 MATHEMATICS AND COMPUTING↗

NASA Operational Simulator for Small Satellites: Tools for Software Based Validation and Verification of Small Satellites

The NASA Operational Simulator for Small Satellites (NOS3) is a suite of tools to aid in areas such as software development, integration test (IT), mission operations training, verification and validation (VV), and software systems check-out. NOS3 provides a software development environment, a multi-target build system, an operator interface-ground station, dynamics and environment simulations, and software-based hardware models. NOS3 enables the development of flight software (FSW) early in the project life cycle, when access to hardware is typically not available. For small satellites there are extensive lead times on many of the commercial-off-the-shelf (COTS) components as well as limited funding for engineering test units (ETU). Considering the difficulty of providing a hardware test-bed to each developer tester, hardware models are modeled based upon characteristic data or manufacturers data sheets for each individual component. The fidelity of each hardware models is such that FSW executes unaware that physical hardware is not present. This allows binaries to be compiled for both the simulation environment, and the flight computer, without changing the FSW source code. For hardware models that provide data dependent on the environment, such as a GPS receiver or magnetometer, an open-source tool from NASA GSFC (42 Spacecraft Simulation) is used to provide the necessary data. The underlying infrastructure used to transfer messages between FSW and the hardware models can also be used to monitor, intercept, and inject messages, which has proven to be beneficial for VV of larger missions such as James Webb Space Telescope (JWST). As hardware is procured, drivers can be added to the environment to enable hardware-in-the-loop (HWIL) testing. When strict time synchronization is not vital, any number of combinations of hardware components and software-based models can be tested. The open-source operator interface used in NOS3 is COSMOS from Ball Aerospace. For testing, plug-ins are implemented in COSMOS to control the NOS3 simulations, while the command and telemetry tools available in COSMOS are used to communicate with FSW. NOS3 is actively being used for FSW development and component testing of the Simulation-to-Flight 1 (STF-1) CubeSat. As NOS3 matures, hardware models have been added for common CubeSat components such as Novatel GPS receivers, ClydeSpace electrical power systems and batteries, ISISpace antenna systems, etc. In the future, NASA IVV plans to distribute NOS3 to other CubeSat developers and release the suite to the open-source community.

Verification↗

Extending C++ for Heterogeneous Quantum-Classical Computing

In this report we present qcor - a language extension to C++ and compiler implementation that enables heterogeneous quantum-classical programming, compilation, and execution in a single-source context. Our work provides a first-of-its-kind C++ compiler enabling high-level quantum kernel (function) expression in a quantum-language agnostic manner, as well as a hardware-agnostic, retargetable compiler workflow targeting a number of physical and virtual quantum computing backends. qcor leverages novel Clang plugin interfaces and builds upon the XACC system-level quantum programming framework to provide a state-of-the-art integration mechanism for quantum-classical compilation that leverages the best from the community at-large. qcor translates quantum kernels ultimately to the XACC intermediate representation, and provides user-extensible hooks for quantum compilation routines like circuit optimization, analysis, and placement. This work details the overall architecture and compiler workflow for qcor, and provides a number of illuminating programming examples demonstrating its utility for near-term variational tasks, quantum algorithm expression, and feed-forward error correction schemes.

97 MATHEMATICS AND COMPUTING↗

Conceptual definition of Automated Power Systems Management

Automated Power Systems Management (APSM) is defined as the capability of a spacecraft power system to automatically perform monitoring, computational, command, and control functions without ground intervention. Power systems for future planetary spacecraft must have this capability because they must perform up to 10 years, and accommodate real-time changes in mission execution autonomously. Specific APSM functions include fault detection, isolation, and correction; system performance and load profile prediction; power system optimization; system checkout; and data storage and transmission control. This paper describes the basic method of implementing these specific functions. The APSM hardware includes a central power system computer and a processor dedicated to each major power system subassembly along with digital interface circuitry. The major payoffs anticipated are in enhancement of spacecraft reliability and life and reduction of overall spacecraft program cost.

Imamura, M. S.↗

Basic Research Needs in Quantum Computing and Networking

Employing quantum mechanical resources in computing, information processing, and networking opens the door to potential exponential advantages over classical counterparts. However, quantifying and realizing such advantages poses extensive scientific and engineering challenges. Department of Energy (DOE) investments have driven steady progress in addressing such challenges. Recently developed quantum algorithms offer asymptotic exponential advantages in speed or accuracy for fundamental scientific problems. These problems include simulating physical systems, solving systems of linear equations, differential equations, and optimization problems. Empirical demonstrations on nascent quantum hardware suggest better performance on contrived computational tasks than classical analogs. However, the requirements for a quantum computer or network to demonstrate an end-to-end rigorously quantifiable performance improvement over classical analogs remains a grand challenge, especially for problems of practical value. In particular, what will be required for quantum technology to ultimately exhibit scalable, rigorous, and transformative performance advantages for practical applications? In July 2023, DOE’s Advanced Scientific Computing Research program in the Office of Science convened the Workshop on Basic Research Needs in Quantum Computing and Networking, where major opportunities and grand challenges were identified. The following five priority research directions (PRDs) were identified as a result of the workshop.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Investigating Formal Methods Tools and their Applicability for Hardware Vulnerability Remediation

Formal methods use mathematical logic and equations to prove that a system or code is secure. In this poster, I examine existing formal methods tools and their application for projects working to remediate vulnerabilities in hardware and hardware description language. This poster is focused on the tools ReWire and AutoGenILA and I hope to evaluate their benefits and weaknesses with the intention of creating an internal report on the application and weaknesses of existing formal methods tools and identifying gaps for future formal methods tool creation.

97 - MATHEMATICS AND COMPUTING↗