SEARCH · Engineering Papers
Results for “Computer Systems”
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.
Battery system management through non-linear estimation of battery state of charge
Systems, methods, and computer media for battery system management and non-linear estimation of battery state of charge are provided herein. Battery data is received for a time period over which a battery system has operated. The battery data represents the actual performance of the battery system over the time period. Sub-periods of charging or discharging can be identified in the time period. For the sub-periods of time, a curve can be fit to the battery data. Using the curves for the battery data for the sub-periods of time, an expected performance of the battery system, over a range of states-of-charge, can be determined. Operating instructions for the battery system can be provided based on the expected performance.
Scalability under a Power Bound using the GREMLINs Framework
With the move towards exascale, system and software developers will have to deal with issues of extreme parallelism. The system properties affected most by the increase in node and core count are the shared resources on node and across the system. The increase in parallelism leads to reduced memory and bandwidth when regarding individual cores. Since power is a limiting factor for supercomputers, and power is not fully utilized in current systems, overprovisioning compute resources is a viable approach to maximized power utilization. To maximize system performance in regard to these changing conditions, it is necessary to understand how resource restrictions impact performance and system behavior. For the purpose of understanding anticipated system properties the GREMLINs Framework was developed. The framework gives the opportunity to add power restrictions, hinder memory properties and introduce faults to study resilience, among others. These features give the opportunity to use current petascale technology to study problems system designers and software developers will have to face when moving towards exascale and beyond. This work describes the initial release of the GREMLINs Framework, developed for this work, and shows how it can be used to study the scaling behavior of proxy applications. These proxy applications represent a selection of HPC workloads important to the scientific community. The proxy applications studied are AMG2013, an algebraic multi-grid linear system solver, CoMD, a classical molecular dynamics proxy application and NEKBONE, an application that uses a high order spectral element method to solve the Navier-Stokes equations. The main interest of these studies lies in analysis regarding their power behavior at scale under a power bound. These findings show how the GREMLINs Framework can help systems and software designers to attain better application performance and can also be used as basis for CPU power balancing tools to use power more efficiently.
Active Learning for Metamaterial Optimization on HPC and QC Integrated Systems
Active learning algorithms, integrating machine learning, quantum computing and optics simulation in an iterative loop, offer a promising approach to optimizing metamaterials. However, these algorithms can face difficulties in optimizing highly complex structures due to computational limitations. High-performance computing (HPC) and quantum computing (QC) integrated systems can address these issues by enabling parallel computing. In this study, we develop an active learning algorithm working on HPC-QC integrated systems. We evaluate the performance of optimization processes within active learning (i.e., training a machine learning model, problem-solving with quantum computing, and evaluating optical properties through wave-optics simulation) for highly complex metamaterial cases. Our results showcase that utilizing multiple cores on the integrated system can significantly reduce computational time, thereby enhancing the efficiency of optimization processes. Therefore, we expect that leveraging HPC-QC integrated systems helps effectively tackle large-scale optimization challenges in general.
A Survey of Distributed Optimization and Control Algorithms for Electric Power Systems
Historically, centrally computed algorithms have been the primary means of power system optimization and control. With increasing penetrations of distributed energy resources requiring optimization and control of power systems with many controllable devices, distributed algorithms have been the subject of significant research interest. Here, this paper surveys the literature of distributed algorithms with applications to optimization and control of power systems. In particular, this paper reviews distributed algorithms for offline solution of optimal power flow (OPF) problems as well as online algorithms for real-time solution of OPF, optimal frequency control, optimal voltage control, and optimal wide-area control problems.
Active Measurement of the Impact of Network SwitchUtilization on Application Performance
Inter-node networks are a key capability of High-Performance Computing (HPC) systems that differentiates them from less capable classes of machines. However, in spite of their very high performance, the increasing computational power of HPC compute nodes and the associated rise in application communication needs make network performance a common performance bottleneck. To achieve high performance in spite of network limitations application developers require tools to measure their applications’ network utilization and inform them about how the network’s communication capacity relates to the performance of their applications. This paper presents a new performance measurement and analysis methodology based on empirical measurements of network behavior. Our approach uses two benchmarks that inject extra network communication. The first probes the fraction of the network that is utilized by a software component (an application or an individual task) to determine the existence and severity of network contention. The second aggressively injects network traffic while a software component runs to evaluate its performance on less capable networks or when it shares the network with other software components. We then combine the information from the two types of experiment to predict the performance slowdown experienced by multiple software components (e.g. multiple processes of a single MPI application) when they share a single network. Our methodology is applied to individual network switches and demonstrated on two real systems with different types of switches and compute node architectures.
Fast, scalable and accurate finite-element based ab initio calculations using mixed precision computing: 46 PFLOPS simulation of a metallic dislocation system
Explore the source record for details and available documents.
A Self-Sustained CPS Design for Reliable Wildfire Monitoring
Continuous monitoring of areas nearby the electric grid is critical for preventing and early detection of devastating wildfires. Existing wildfire monitoring systems are intermittent and oblivious to local ambient risk factors, resulting in poor wildfire awareness. Ambient sensor suites deployed near the gridlines can increase the monitoring granularity and detection accuracy. However, these sensors must address two challenging and competing objectives at the same time. First, they must remain powered for years without manual maintenance due to their remote locations. Second, they must provide and transmit reliable information if and when a wildfire starts. The first objective requires aggressive energy savings and ambient energy harvesting, while the second requires continuous operation of a range of sensors. To the best of our knowledge, this paper presents the first self-sustained cyber-physical system that dynamically co-optimizes the wildfire detection accuracy and active time of sensors. The proposed approach employs reinforcement learning to train a policy that controls the sensor operations as a function of the environment (i.e., current sensor readings), harvested energy, and battery level. Here, the proposed cyber-physical system is evaluated extensively using real-life temperature, wind, and solar energy harvesting datasets and an open-source wildfire simulator. In long-term (5 years) evaluations, the proposed framework achieves 89% uptime, which is 46% higher than a carefully tuned heuristic approach. At the same time, it averages a 2-minute initial response time, which is at least 2.5× faster than the same heuristic approach. Furthermore, the policy network consumes 0.6 mJ per day on the TI CC2652R microcontroller using TensorFlow Lite for Micro, which is negligible compared to the daily sensor suite energy consumption.
Sub-MeV ultra-compact linac for radioactive isotope sources replacement, non-destructive testing, security and medical applications
RadiaBeam Technologies, LLC has developed an inexpensive, hand-portable 180 keV electron accelerator to replace 57 Co radionuclide sources in Cascade Header Enrichment Monitor detectors. We used two innovative technologies in our design: a Ku-band (1 cm wavelength) magnetron, and a split accelerating structure design to reduce manufacturing costs. In this paper, we will discuss the accelerator, including X-ray convertor and accelerating structure design. The results of RF measurements of a Ku-band split structure prototype will also be reviewed. Other applications of Ku-band linacs include compact backscatter and transmission X-ray security inspection systems, as well as computed tomography (CT) systems for luggage and parcel screening. The main requirement for luggage inspection systems is that the X-ray source must be extremely compact. Further, the required energies should cover the range from 180 keV to 1 MeV. In this paper, we will discuss the linac’s energy scalability and other modifications that will enable our technology to address these applications.
Modeling Electronic Quantum Transport with Machine Learning
We present a machine learning approach to solve electronic quantum transport equations of one-dimensional nanostructures. The transmission coefficients of disordered systems were computed to provide training and test data sets to the machine. The system’s representation encodes energetic as well as geometrical information to characterize similarities between disordered configurations, while the Euclidean norm is used as a measure of similarity. Errors for out-of-sample predictions systematically decrease with training set size, enabling the accurate and fast prediction of new transmission coefficients. The remarkable performance of our model to capture the complexity of interference phenomena lends further support to its viability in dealing with transport problems of undulatory nature.
Singularity v1.x
Singularity is a container solution designed to facilitate mobility of compute across systems and HPC infrastructures. It does this by creating minimal containers that are defined by a specfile and files from the host system are used to build the container. The resulting container can then be launched by any Linux computer with Singularity installed regardless if the programs inside the container are present on the target system, or if they are a different version, or even incompatible versions. Singularity achieves extreme portability without sacrificing usability thus solving the need of mobility of compute. Singularity containers can be executed within a normal/standard command line process flow.
Proceedings of International Symposium on Sensor Networks, Systems and Security: Advances in Computing and Networking with Applications
This state of the art book discusses current trends in sensor networks, systems, and security. This proceedings contains a collection of articles authored by outstanding researchers in the field, providing an excellent reference for graduate students and researchers in this area.
iez: Resource Contention Aware Load Balancing for Large-Scale Parallel File Systems
Parallel I/O performance is crucial to sustaining scientific applications on large-scale High-Performance Computing (HPC) systems. However, I/O load imbalance in the underlying distributed and shared storage systems can significantly reduce overall application performance. There are two conflicting challenges to mitigate this load imbalance: (i) optimizing systemwide data placement to maximize the bandwidth advantages of distributed storage servers, i.e., allocating I/O resources efficiently across applications and job runs; and (ii) optimizing client-centric data movement to minimize I/O load request latency between clients and servers, i.e., allocating I/O resources efficiently in service to a single application and job run. Moreover, existing approaches that require application changes limit wide-spread adoption in commercial or proprietary deployments. We propose iez, an “end-to-end control plane” where clients transparently and adaptively write to a set of selected I/O servers to achieve balanced data placement. Our control plane leverages realtime load information for distributed storage server global data placement while our design model leverages trace-based optimization techniques to minimize I/O load request latency between clients and servers. We evaluate our proposed system on an experimental cluster for two common use cases: synthetic I/O benchmark IOR for large sequential writes and a scientific application I/O kernel, HACC-I/O. Results show read and write performance improvements of up to 34% and 32%, respectively, compared to the state of the art.
Demonstration of Intelligent Control and Fan Improvements in Computer Room Air Handlers
This report documents a demonstration of the energy-efficiency improvement provided by a new control system for computer room air handling devices. It also analyzes measured and reported air handling device fan power associated with changing the fan type. A 135,000 square foot commercial data center was used for the demonstration. All air handling units were upgraded with improved efficiency fans, and a control system that automatically adjusts the fan speed for the air handling units was added. Power measurements were collected for a baseline and for a period with the fan speed control system active. Changing the fan type resulted in a savings of 47 percent of energy used by the air handling equipment and associated chiller plant energy needed to cool the air handlers themselves. The addition of the fan speed control resulted in an additional 37 percent savings in the same two categories. The combined savings for the two improvements for the same categories was 66 percent compared to the data center fitted with the original fans without a control system. The energy use reduction provided by the complete air handling device improvement program for the whole data center site is estimated to be 2.9 million kilowatt hours per year—an overall data center site savings of 8.0 percent. The reduced electrical energy use at the site provides a 1.9 million pound yearly reduction of carbon dioxide emissions. This demonstration showed that fan upgrades and a control system addition provide cost-effective improvements for data centers, with a payback reported to be under two years without utility incentives. In addition to the control system providing energy savings, the data collection and visual analysis capabilities provided immediate and long-term benefits. It is recommended that data center operators consider investing in fan upgrades and/or adding fan speed control for computer room air handlers.
Quantum Monte Carlo calculations in configuration space with three-nucleon forces
Neutron matter, through its connection to neutron stars as well as systems like cold atom gases, is one of the most interesting yet computationally accessible systems in nuclear physics. The Configuration-Interaction Monte Carlo (CIMC) method is a stochastic many-body technique allowing to tackle strongly coupled systems. In contrast to other Quantum Monte Carlo methods employed in nuclear physics, the CIMC method can be formulated directly in momentum space allowing for an efficient use of non-local interactions. In this work we extend CIMC method to include three-nucleon interactions through the normal-ordered two-body approximation. Here, we present results for the equation of state of neutron matter in line with other many-body calculations that employ low resolution chiral interactions, and provide predictions for the momentum distribution and the static structure factor.
UnifyFS: A User-level Shared File System for Unified Access to Distributed Local Storage
We introduce UnifyFS, a user-level file system that aggregates node-local storage tiers available on high performance computing (HPC) systems and makes them available to HPC applications under a unified namespace. UnifyFS employs transparent I/O interception, so it does not require changes to application code and is compatible with commonly used HPC I/O libraries. The design of UnifyFS supports the predominant HPC I/O workloads and is optimized for bulk-synchronous I/O patterns. Furthermore, UnifyFS provides customizable file system semantics to flexibly adapt its behavior for diverse I/O workloads and storage devices. In this paper, we discuss the unique design goals and architecture of UnifyFS and evaluate its performance on a leadership-class HPC system. In our experimental results, we demonstrate that UnifyFS exhibits excellent scaling performance for write operations and can improve the performance of application checkpoint operations by as much as 3× versus a tuned configuration.
Design and Analysis of Digital Communication Within an SoC-Based Control System for Trapped-Ion Quantum Computing
Not Available
INTEGRATION OF PARTICLE-GAS SYSTEMS WITH STIFF MUTUAL DRAG INTERACTION
Numerical simulation of numerous mm/cm-sized particles embedded in a gaseous disk has become an important tool in the study of planet formation and in understanding the dust distribution in observed protoplanetary disks. However, the mutual drag force between the gas and the particles can become so stiff—particularly because of small particles and/or strong local solid concentration—that an explicit integration of this system is computationally formidable. In this work, we consider the integration of the mutual drag force in a system of Eulerian gas and Lagrangian solid particles. Despite the entanglement between the gas and the particles under the particle-mesh construct, we are able to devise a numerical algorithm that effectively decomposes the globally coupled system of equations for the mutual drag force, and makes it possible to integrate this system on a cell-by-cell basis, which considerably reduces the computational task required. We use an analytical solution for the temporal evolution of each cell to relieve the time-step constraint posed by the mutual drag force, as well as to achieve the highest degree of accuracy. To validate our algorithm, we use an extensive suite of benchmarks with known solutions in one, two, and three dimensions, including the linear growth and the nonlinear saturation of the streaming instability. We demonstrate numerical convergence and satisfactory consistency in all cases. Our algorithm can, for example, be applied to model the evolution of the streaming instability with mm/cm-sized pebbles at high mass loading, which has important consequences for the formation scenarios of planetesimals.