Search NASA⌕ Search

SEARCH · Search NASA

Results for “hardware 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 901 records · Page 50

Chisel: reliability- and accuracy-aware optimization of approximate computational kernels

The accuracy of an approximate computation is the distance between the result that the computation produces and the corresponding fully accurate result. The reliability of the computation is the probability that it will produce an acceptably accurate result. Emerging approximate hardware platforms provide approximate operations that, in return for reduced energy consumption and/or increased performance, exhibit reduced reliability and/or accuracy. We present Chisel, a system for reliability- and accuracy-aware optimization of approximate computational kernels that run on approximate hardware platforms. Given a combined reliability and/or accuracy specification, Chisel automatically selects approximate kernel operations to synthesize an approximate computation that minimizes energy consumption while satisfying its reliability and accuracy specification. We evaluate Chisel on five applications from the image processing, scientific computing, and financial analysis domains. The experimental results show that our implemented optimization algorithm enables Chisel to optimize our set of benchmark kernels to obtain energy savings from 8.7% to 19.8% compared to the fully reliable kernel implementations while preserving important reliability guarantees.

97 MATHEMATICS AND COMPUTING↗

Hardware Evaluation Outreach: Application development Challenges Now and for the Exascale Era

The intent of this document is to assist the programmer in understanding details of contemporary and Exascale hardware system design and how these designs provide opportunities and place constraints on next-generation simulation software design. We attempt to clarify hardware organization and component details for our most current and Exascale systems to help program developers understand how software needs to change in order to take best advantage of the performance available. Exascale success is specifically defined for ECP as a 50x improvement over baseline in the aggregate "capability volume" on several KPP axes, of which raw floating point performance is only one, but also includes characteristics such as problem size, system memory size, node memory size, power, and efficiency. This multi-axis approach is particularly important to understand in the context of delivered improvements in real applications, since, for instance, the floating point computation may comprise less than 10% of the actual computational work required. Given the Exascale requirements and the constraints these requirements put on the performance expectations of fundamental system components, the programmer will be forced to re-think several application implementation details in order to achieve exaflop performance on these systems. The remainder of this document aims to present more detail on Exascale era system hardware and the specific areas that the programmer should address to extract performance from these systems. We attempt to give the programmer guidance at both a high- and low-level, providing some abstract suggestions on how to refactor codes given the expected system architectures and some low-level recommendations on how to implement these modifications. We also include a section on training resources that are helpful for both programmers that are just beginning to understand code modifications for contemporary and Exascale systems and for those that have done some refactoring and are now trying to extract maximal application performance from these systems.

97 MATHEMATICS AND COMPUTING↗

Distributionally Robust Variational Quantum Algorithms With Shifted Noise

Given their potential to demonstrate near-term quantum advantage, variational quantum algorithms (VQAs) have been extensively studied. Although numerous techniques have been developed for VQA parameter optimization, it remains a significant challenge. A practical issue is the high sensitivity of quantum noise to environmental changes, and its propensity to shift in real time. This presents a critical problem as an optimized VQA ansatz may not perform effectively under a different noise environment. For the first time, we explore how to optimize VQA parameters to be robust against unknown shifted noise. We model the noise level as a random variable with an unknown probability density function (PDF), and we assume that the PDF may shift within an uncertainty set. This assumption guides us to formulate a distributionally robust optimization problem, with the goal of finding parameters that maintain effectiveness under shifted noise. We utilize a distributionally robust Bayesian optimization solver for our proposed formulation. This provides numerical evidence in both the Quantum Approximate Optimization Algorithm (QAOA) and the Variational Quantum Eigensolver (VQE) with hardware-efficient ansatz, indicating that we can identify parameters that perform more robustly under shifted noise. We regard this work as the first step towards improving the reliability of VQAs influenced by real-time noise.

97 MATHEMATICS AND COMPUTING↗

Libpanda: A High Performance Library for Vehicle Data Collection

Cyber-Physical Systems (CPS) generally involve time-critical components due to physical dynamics, therefore necessitating high-performance subsystems. This is also true in data collection scenarios to infer physical phenomena. This paper covers Libpanda as an example of a component that has been designed to address performance issues in CPS implementations. Libpanda is a C++ library that interfaces software with a Comma.ai Panda device. Pandas are used for installation in modern vehicles to read the vehicle CAN bus, providing rich sensor data and limited vehicle control through message injection. The motivation to design lib-panda stems from the lack of performance in Python-based code that runs on inexpensive hardware like a Raspberry Pi. In such situations, Python code would result in utilizing 92% CPU while also dropping around 40% of the CAN packet due to bottlenecks. Without using different tools, inconsistent data collection means a loss of time-based vehicle state interpretation. Libpanda addresses these issues through implementation in a different language and implementation of different design paradigms involving asynchronous calls and multithreading. The Panda also features a GPS module that allows multiple instances to synchronize clocks for large-scale data collection scenarios. Libpanda has been designed with time-synchronization in mind to aid in the measurement of inter-vehicle dynamics. The performance improvements of libpanda have resulted in it becoming an important component in automotive dynamics research that requires a higher technical performance in large-scale experiments.

Bunting, Matthew↗

Aho-Corasick String Matching on Shared and Distributed Memory Parallel Architectures

String matching is at the core of many critical applications, including network intrusion detection systems, search engines, virus scanners, spam filters, DNA and protein sequencing, and data mining. For all of these applications string matching requires a combination of (sometimes all) the following characteristics: high and/or predictable performance, support for large data sets and flexibility of integration and customization. Many software based implementations targeting conventional cache-based microprocessors fail to achieve high and predictable performance requirements, while Field-Programmable Gate Array (FPGA) implementations and dedicated hardware solutions fail to support large data sets (dictionary sizes) and are difficult to integrate and customize. The advent of multicore, multithreaded, and GPU-based systems is opening the possibility for software based solutions to reach very high performance at a sustained rate. This paper compares several software-based implementations of the Aho-Corasick string searching algorithm for high performance systems. We discuss the implementation of the algorithm on several types of shared-memory high-performance architectures (Niagara 2, large x86 SMPs and Cray XMT), distributed memory with homogeneous processing elements (InfiniBand cluster of x86 multicores) and heterogeneous processing elements (InfiniBand cluster of x86 multicores with NVIDIA Tesla C10 GPUs). We describe in detail how each solution achieves the objectives of supporting large dictionaries, sustaining high performance, and enabling customization and flexibility using various data sets.

99 GENERAL AND MISCELLANEOUS↗

OPET Hardware (Open PV Electrical Tool Hardware) [SWR-25-42]

OPET (Open-source Photovoltaic Electrical Tool) is used for performance measurements of solar photovoltaic (PV) devices in the field under natural sunlight or in the lab under artificial light. Its primary use is in research and development of solar cells and modules, specifically in reliability and durability research of PV devices. Some features and functions include: -IV curve measurements with linear or cosine distributed measurement points -PV device active loading at open circuit voltage (Voc), short circuit current (Isc) and maximum power point (Pmp) -Bias power supply to overcome series resistance in contact wires for Isc measurements and loading -PV voltage input in five ranges from 1V to 100V -PV current input ranges -Low current version, six current ranges from 1.1mA to 340mA -High current version, six current ranges from 50mA to 15A -IO ports for I2C and SPI temperature sensor Arduino extension boards -Integrated fan control This repository contains everything relating to the hardware of the OPET device. If you are looking for the firmware or software repositories, links are below: https://github.com/NREL/opet-firmware https://github.com/NREL/opet-control

McDanold, Byron [National Renewable Energy Laborat↗

Efficient Mixed-Precision Matrix Factorization of the Inverse Overlap Matrix in Electronic Structure Calculations with AI-Hardware and GPUs

In recent years, a new kind of accelerated hardware has gained popularity in the artificial intelligence (AI) community which enables extremely high-performance tensor contractions in reduced precision for deep neural network calculations. In this article, we exploit Nvidia Tensor cores, a prototypical example of such AI-hardware, to develop a mixed precision approach for computing a dense matrix factorization of the inverse overlap matrix in electronic structure theory, S –1 . This factorization of S –1 , written as ZZT = S –1 , is used to transform the general matrix eigenvalue problem into a standard matrix eigenvalue problem. Here we present a mixed precision iterative refinement algorithm where Z is given recursively using matrix–matrix multiplications and can be computed with high performance on Tensor cores. To understand the performance and accuracy of Tensor cores, comparisons are made to GPU-only implementations in single and double precision. Additionally, we propose a nonparametric stopping criteria which is robust in the face of lower precision floating point operations. The algorithm is particularly useful when we have a good initial guess to Z, for example, from previous time steps in quantum-mechanical molecular dynamics simulations or from a previous iteration in a geometry optimization.

36 MATERIALS SCIENCE↗

A Runtime Environment for Supporting Research in Resilient HPC System Software & Tools

The high-performance computing (HPC) community continues to increase the size and complexity of hardware platforms that support advanced scientific workloads. The runtime environment (RTE) is a crucial layer in the software stack for these large-scale systems. The RTE manages the interface between the operating system and the application running in parallel on the machine. The deployment of applications and tools on large-scale HPC computing systems requires the RTE to manage process creation in a scalable manner, support sparse connectivity, and provide fault tolerance. We have developed a new RTE that provides a basis for building distributed execution environments and developing tools for HPC to aid research in system software and resilience. This paper describes the software architecture of the Scalable runTime Component Infrastructure (STCI), which is intended to provide a complete infrastructure for scalable start-up and management of many processes in large-scale HPC systems. We highlight features of the current implementation, which is provided as a system library that allows developers to easily use and integrate STCI in their tools and/or applications. The motivation for this work has been to support ongoing research activities in fault-tolerance for large-scale systems. We discuss the advantages of the modular framework employed and describe two use cases that demonstrate its capabilities: (i) an alternate runtime for a Message Passing Interface (MPI) stack, and (ii) a distributed control and communication substrate for a fault-injection tool.

Computer Science↗

Hybrid Oscillator-Qubit Quantum Processors: Instruction Set Architectures, Abstract Machine Models, and Applications

This tutorial offers a pedagogical guide to hybrid quantum processors that integrate discrete-variable (DV) qubits and continuous-variable (CV) oscillators. Aimed at computer scientists, engineers, and physicists, it provides an overview of the experimental, algorithmic, and architectural aspects of this novel and rapidly developing hardware model. Experimental realizations of this model include superconducting, trapped-ion, and neutral-atom platforms. By combining DV and CV components, hybrid oscillator-qubit processors enable a powerful new paradigm that offers complementary strengths for quantum control, error correction, computation, and simulation. Working toward the goal of a full-stack system connecting applications to CV-DV hardware, we define and formulate abstract machine models and instruction set architectures. These essential abstractions enable codesign of hardware and software, and resource estimation for exploring the potential of current and future hardware for computational and simulation tasks. Using these abstractions, we present both new and existing examples that illustrate the benefits of hybrid CV-DV processors relative to traditional DV-only hardware in computation as well as quantum simulation of physical models. Examples include algorithms for transferring states between DV and CV systems, performing the quantum Fourier transform, and simulation of lattice gauge theories. Relative to qubit-only hardware, the bosonic degrees of freedom natively available in hybrid architectures can substantially reduce the circuit complexity of simulations for physical models containing bosons. A key technique is the extension of quantum signal processing ideas to CV-DV systems. This work is intended to serve as a timely and comprehensive guide to this relatively unexplored yet promising approach to quantum computation and to provide a road map to guide future development.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Efficient, Compact, and Smooth Variable Propulsion Motor (Final Report)

In this project, a new architecture of highly efficient hydraulic motor was developed for the propulsion of off-highway vehicles. The motor uses an adjustable linkage driving a cam to vary the displacement of the piston, resulting in a Variable Displacement Linkage Motor (VDLM). The motor uses low friction rolling element bearings to significantly reduce mechanical friction, especially in the demanding low-speed high-torque conditions experienced by off-highway vehicles. The VDLM has high torque capabilities for its size due to the radial piston packaging and use of a multi-lobe cam. A VDLM is very smooth due to the ability to tune the torque ripple through the design of the cam profile. The project was divided into three periods. During the first period, a dynamic model was constructed of the motor to predict the performance of the motor and the vehicle. During the second period, a single-cylinder learning prototype was designed, built, and tested to validate the models constructed in the first period. In the third period, a multi-cylinder prototype motor was optimized, designed, fabricated, and tested. The motor demonstrated excellent mechanical efficiency (above 92.5% across the range of displacements), but the experimentally measure volumetric efficiency was lower than expected due to higher leakage rates created by the poor tolerance control on the prototype. To validate the dynamic models developed in the first period and better understand design trade-offs. In the third period a multi-cylinder concept demonstration prototype will be designed, fabricated, and tested. The final prototype will be tested on a motor dynamometer and will be utilized in hardware-in-the-loop testing to demonstrate its efficiency and performance impacts on the overall drive train. The experimental results were used in a drive train simulation of a compact track loader operating through a drive cycle. Using the VDLM in a hydrostatic circuit yielded 17.1% reduction in fuel consumption and 36.5% reduction in a series hybrid transmission.

99 GENERAL AND MISCELLANEOUS↗

Improving Large-scale Storage System Performance via Topology-aware and Balanced Data Placement

With the advent of big data, the I/O subsystems of large-scale compute clusters are becoming a center of focus, with more applications putting greater demands on end-to-end I/O performance. These subsystems are often complex in design. They comprise of multiple hardware and software layers to cope with the increasing capacity, capability and scalability requirements of data intensive applications. The sharing nature of storage resources and the intrinsic interactions across these layers make it to realize user-level, end-to-end performance gains a great challenge. We propose a topology-aware resource load balancing strategy to improve per-application I/O performance. We demonstrate the effectiveness of our algorithm on an extreme-scale compute cluster, Titan, at the Oak Ridge Leadership Computing Facility (OLCF). Our experiments with both synthetic benchmarks and a real-world application show that, even under congestion, our proposed algorithm can improve large-scale application I/O performance significantly, resulting in both the reduction of application run times and higher resolution simulation runs.

Wang, Feiyi↗

Detecting and Blocking Network Attacks at Ultra High Speeds

Stateful, in-depth, in-line traffic analysis for intrusion detection and prevention has grown increasingly more difficult as the data rates of modern networks rise. One point in the design space for high-performance network analysis - pursued by a number of commercial products - is the use of sophisticated custom hardware. For very high-speed processing, such systems often cast the entire analysis process in ASICs. This project pursued a different architectural approach, which we term Shunting. Shunting marries a conceptually quite simple hardware device with an Intrusion Prevention System (IPS) running on commodity PC hardware. The overall design goal is was to keep the hardware both cheap and readily scalable to future higher speeds, yet also retain the unparalleled flexibility that running the main IPS analysis in a full general-computing environment provides. The Shunting architecture we developed uses a simple in-line hardware element that maintains several large state tables indexed by packet header fields, including IP/TCP flags, source and destination IP addresses, and connection tuples. The tables yield decision values the element makes on a packet-by-packet basis: forward the packet, drop it, or divert ('shunt') it through the IPS (the default). By manipulating table entries, the IPS can, on a fine-grained basis: (i) specify the traffic it wishes to examine, (ii) directly block malicious traffic, and (iii) 'cut through' traffic streams once it has had an opportunity to 'vet' them, or (iv) skip over large items within a stream before proceeding to further analyze it. For the Shunting architecture to yield benefits, it needs to operate in an environment for which the monitored network traffic has the property that - after proper vetting - much of it can be safely skipped. This property does not universally hold. For example, if a bank needs to examine all Web traffic involving its servers for regulatory compliance, then a monitor in front of one of the bank's server farms cannot safely omit a subset of the traffic from analysis. In this environment, Shunting cannot realize its main performance benefits, and the monitoring task likely calls for using custom hardware instead. However, in many other environments we find Shunting holds promise for delivering major performance gains. This arises due to the the widely documented 'heavy tail' nature of most forms of network traffic, which we might express as 'a few of the connections carry just about all the bytes.' The key additional insight is '... and very often for these few large connections, the very beginning of the connection contains nearly all the information of interest from a security analysis perspective.' We argue that this second claim holds because it is at the beginning of connections that authentication exchanges occur, data or file names and types are specified, request and reply status codes conveyed, and encryption is negotiated. Once these occur, we have seen most of the interesting facets of the dialog. Certainly the remainder of the connection might also yield some grist for analysis, but this is generally less likely, and thus if we want to lower analysis load at as small a loss as possible of information relevant to security analysis, we might best do so by skipping the bulk of large connections. In a different context, the 'Time Machine' work by Kornexl and colleagues likewise shows that in some environments we can realize major reductions in the volume of network traffic processed, by limiting the processing to the first 10-20 KB of each connection. As a concrete example, consider an IPS that monitors SSH traffic. When a new SSH connection arrives and the Shunt fails to find an entry for it in any of its tables (per-address, per-port, per-connection), it executes the default action of diverting the connection through the IPS. The IPS analyzes the beginning of the connection in this fashion. As long as it is satisified with the dialog, it reinjects the packets forwarded to it so that the connection can continue. If the connection successfully negotiates encryption, the IPS can no longer profitably analyze it, so it downloads a per-connection table entry to the Shunt specifying that the action for the connection in the future is 'forward.' For heavy-tailed connections, this means a very large majority of the connection's packets will now pass through the Shunt device without burdening the IPS with any further analysis load. On the other hand, if the IPS is dissatisfied with some element of the initial dialog, it downloads a 'drop' entry to terminate the connection. Note that by providing for reinjection, we can promote an intrusion detection system into an intrusion prevention system, one that does not merely detect attacks but can block them before they complete. Reinjection also allows the IPS to normalize traffic to remove ambiguities that attackers can leverage to evade the IPS.

99 GENERAL AND MISCELLANEOUS↗

Examining System-Wide Impacts of Solar PV Control Systems with a Power Hardware-in-the-Loop Platform

High penetration levels of distributed solar PV power generation can lead to adverse power quality impacts such as excessive voltage rise, voltage flicker, and reactive power values that result in unacceptable voltage levels. Advanced inverter control schemes have been proposed that have the potential to mitigate many power quality concerns. However, closed-loop control may lead to unintended behavior in deployed systems as complex interactions can occur between numerous operating devices. In order to enable the study of the performance of advanced control schemes in a detailed distribution system environment, a Hardware-in-the-Loop (HIL) platform has been developed. In the HIL system, GridLAB-D, a distribution system simulation tool, runs in real-time mode at the Pacific Northwest National Laboratory (PNNL) and supplies power system parameters at a point of common coupling to hardware located at the National Renewable Energy Laboratory (NREL). Hardware inverters interact with grid and PV simulators emulating an operational distribution system and power output from the inverters is measured and sent to PNNL to update the real-time distribution system simulation. The platform is described and initial test cases are presented. The platform is used to study the system-wide impacts and the interactions of controls applied to inverters that are integrated into a simulation of the IEEE 8500-node test feeder, with inverters in either constant power factor control or active volt/VAR control. We demonstrate that this HIL platform is well-suited to the study of advanced inverter controls and their impacts on the power quality of a distribution feeder. Additionally, the results from HIL are used to validate GridLAB-D simulations of advanced inverter controls.

14 SOLAR ENERGY↗

Improving large-scale storage system performance via topology-aware and balanced data placement

With the advent of big data, the I/O subsystems of large-scale compute clusters are becoming a center of focus. More applications are putting greater demands on end-to-end I/O performance. These subsystems are often complex in design. They comprise of multiple hardware and software layers to cope with the increasing capacity, capability, and scalability requirements of data intensive applications. However, the sharing nature of storage resources and the intrinsic interactions across these layers make it a great challenge to realize end-to-end performance gains. This paper proposes a topology-aware strategy to balance the load across resources, to improve the per-application I/O performance. We demonstrate the effectiveness of our algorithm on an extreme-scale compute cluster, Titan, at the Oak Ridge Leadership Computing Facility (OLCF). Our experiments with both synthetic benchmarks and a real-world application show that, even under congestion, our proposed algorithm can improve large-scale application I/O performance significantly, resulting in both a reduction in application run time as well as a higher resolution of simulation run.

Wang, Feiyi↗

Zero and Finite Temperature Quantum Simulations Powered by Quantum Magic

We introduce a quantum information theory-inspired method to improve the characterization of many-body Hamiltonians on near-term quantum devices. We design a new class of similarity transformations that, when applied as a preprocessing step, can substantially simplify a Hamiltonian for subsequent analysis on quantum hardware. By design, these transformations can be identified and applied efficiently using purely classical resources. In practice, these transformations allow us to shorten requisite physical circuit-depths, overcoming constraints imposed by imperfect near-term hardware. Importantly, the quality of our transformations is t u n a b l e : we define a 'ladder' of transformations that yields increasingly simple Hamiltonians at the cost of more classical computation. Using quantum chemistry as a benchmark application, we demonstrate that our protocol leads to significant performance improvements for zero and finite temperature free energy calculations on both digital and analog quantum hardware. Specifically, our energy estimates not only outperform traditional Hartree-Fock solutions, but this performance gap also consistently widens as we tune up the quality of our transformations. In short, our quantum information-based approach opens promising new pathways to realizing useful and feasible quantum chemistry algorithms on near-term hardware.

Physics↗

On the performance and energy efficiency of sparse linear algebra on GPUs

In this paper we unveil some performance and energy efficiency frontiers for sparse computations on GPU-based supercomputers. We compare the resource efficiency of different sparse matrix–vector products (SpMV) taken from libraries such as cuSPARSE and MAGMA for GPU and Intel’s MKL for multicore CPUs, and develop a GPU sparse matrix–matrix product (SpMM) implementation that handles the simultaneous multiplication of a sparse matrix with a set of vectors in block-wise fashion. While a typical sparse computation such as the SpMV reaches only a fraction of the peak of current GPUs, we show that the SpMM succeeds in exceeding the memory-bound limitations of the SpMV. We integrate this kernel into a GPU-accelerated Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) eigensolver. LOBPCG is chosen as a benchmark algorithm for this study as it combines an interesting mix of sparse and dense linear algebra operations that is typical for complex simulation applications, and allows for hardware-aware optimizations. In a detailed analysis we compare the performance and energy efficiency against a multi-threaded CPU counterpart. The reported performance and energy efficiency results are indicative of sparse computations on supercomputers.

Anzt, Hartwig↗

Evolution of Phase Transformation Behavior in Li(Mn1.5Ni0.5)O4 Cathodes Studies By In Situ XRD

Li(Mn1.5Ni0.5)O4 (MNO) is promising spinel cathode material for lithium ion batteries (LIB) due to its increased intercalation potential and improved capacity retention compared to LiMn2O4. Still, improvements to this materials performance must be achieved before MNO is a viable option for commercial LIB cells. In situ X-ray diffraction of Li(Mn1.5Ni0.5)O4 was performed using a novel electrochemical cell based on coin cell hardware. A diffractogram of the pristine material showed a spinel structure with a Ni2+ oxidation state. As the cell was charged through its 4.75V plateau, a transition between spinels with Ni2+, Ni3+, and Ni4+ oxidation was observed. As the oxidation stage on the nickel increased the lattice parameter of the corresponding spinel was reduced. As discharged ensued the spinel reversed its phases change until only the Ni2+ spinel was present. When discharge reached the 2.75V plateau a tetragonal spinel phase was formed, which upon subsequent cell charging was completely converted back to a cubic spinel phase. Lattice parameter changes of each phase were calculated and showed a characteristic strain release during phase changes. After 15 full cycles the transition between these phases was no longer complete and the formation of the tetragonal spinel phase was no longer detected. A discussion of how these cycle-induced changes to phase transition behavior relate to capacity fade and overall cell performance is presented.

74 ATOMIC AND MOLECULAR PHYSICS↗

ZFP Hardware Implementation

As core counts increase in new HPC systems with comparatively little increase in memory bandwidth, the trend is an effective decrease in memory bandwidth per core. Other bandwidth limitations in HPC systems exist between CPU and GPU memory, between system nodes, and between node memory and storage. Compression of floating-point data has the potential to reduce data movement and pressure across these communication channels. Furthermore, it has the potential to reduce the footprint of floating-point arrays stored in memory. ZFP, implemented in software, is gaining traction as an effective method in floating-point compression; however, performance gains are limited to the spare compute cycles available before reaching the bandwidth limitations of the communication channel. A hardware implementation of ZFP has the potential to raise the bar on performance. From the inception of ZFP, it was designed to accommodate a hardware implementation.

42 ENGINEERING↗