Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “PERFORM”

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 1,099 records · Page 61
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Performance Evaluation of Adaptive Routing on Dragonfly-based Production Systems

Performance of applications in production environments can he sensitive to network congestion. Cray Aries supports adaptively routing each network packet independently based on the load or congestion encountered as a packet traverses the network. Software can dictate different routing policies, adjusting between minimal and non-minimal bias, for each posted message. We have extensively evaluated the sensitivity of the routing bias selection on application performance as well as whole system performance in both production and controlled conditions. We show that the default routing bias used in Aries-based systems is often sub-optimal and that using a higher bias towards minimal routes will not only reduce the congestion effects on the application but also will decrease the overall congestion on the network. This routing scheme results in not only improved mean performance (by up to 12%) of most production applications hut also reduced run-to-run variability. Our study prompted the two supercomputing facilities (ALCF and NERSC) to change the default routing mode on their Aries-based systems. We present the substantial improvement measured in the overall congestion management and interconnect performance in production after making this change.

Chunduri, Sudheer↗
DOE OSTI2022

Performance Evaluations of Noisy Approximate Quantum Fourier Arithmetic

The Quantum Fourier Transform (QFT) grants competitive advantages, especially in resource usage and circuit approximation, for performing arithmetic operations on quantum computers, and offers a potential route towards a numerical quantum-computational paradigm. In this paper, we utilize efficient techniques to implement QFT-based integer addition and multiplications. These operations are fundamental to various quantum applications including Shor’s algorithm, weighted sum optimization problems in data processing and machine learning and quantum algorithms requiring inner products. We carry out performance evaluations of these implementations based on IBM’s superconducting qubit architecture using different compatible noise models. We isolate the sensitivity of the component quantum circuits on both one-/two-qubit gate error rates, and the number of the arithmetic operands’ superposed integer states. We analyze performance, and identify the most effective approximation depths for quantum add and quantum multiply within the given context. We observe significant dependency of the optimal approximation depth on the degree of machine noise and the number of superposed states in certain performance regimes. Finally, we elaborate on the algorithmic challenges - relevant to signed, unsigned, modular and non-modular versions - that could also be applied to current implementations of QFT-based subtraction, division, exponentiation, and their potential tensor extensions. Here, we analyze performance trends in our results and speculate on possible future development within this computational paradigm.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Evaluating Unified Memory Performance in HIP

Heterogeneous unified memory management between a CPU and a GPU is a major challenge in GPU computing. Recently, unified memory (UM) has been supported by software and hardware components on AMD computing platforms. The support could simplify the complexities of memory management. In this paper, we attempt to have a better understanding of UM by evaluating the performance of UM programs on an AMD MI100 GPU. More specifically, we evaluate data migration using UM against other data transfer techniques for the overall performance of an application, assess the impacts of three commonly used optimization techniques on the kernel execution time of a vector add sample, and compare the performance and productivity of selected benchmarks with and without UM. The performance overhead associated with UM is not trivial, but it can improve programming productivity by reducing lines of code for scientific applications. We aim to present early results and feedback on the UM performance to the vendor.

Jin, Zheming↗
DOE OSTI2024

IRIS: Exploring Performance Scaling of the Intelligent Runtime System and its Dynamic Scheduling Policies

High-Performance Computing is becoming increasingly heterogeneous, relying on a diverse mix of hardware to achieve good performance. Paradoxically, current drivers and frameworks for these devices typically require separate languages and implementations for each vendor. Furthermore, there are few tools and little support to schedule codes between these devices in a truly heterogeneous manner-partly because of this fragmentation between vendors and the languages each supports. To overcome both limitations, the Intelligent Runtime System (IRIS) was developed. It allows a common task abstraction to automatically be shared among contemporary vendors and is run from a single host-side API. At runtime, IRIS queries the host system and registers which frameworks and drivers are available, these determine which kernels can be used by the scheduler-CPUs via OpenMP, Nvidia GPUs (CUDA), AMD GPUs (HIP), and Intel and Xilinx FPGAs with OpenCL. IRIS enables tasks to be scheduled to any heterogeneous device and resolves to the appropriate kernel binary at runtimeit only uses the devices supported by the system on which it is run. IRIS supports single-task and graph-based expressions of dependencies of tasks. Additionally, IRIS features a range of dynamic scheduling policies, allowing complex chains of tasks and interactions to be executed, relieving the programmer/user from considering the system to assign tasks to devices optimally. This paper presents the peak performance attainable by IRIS over a range of systems-each with different numbers and types of accelerator devices, it highlights the flexibility of IRIS since these devices are truly heterogeneous, relying on different backends (drivers, frameworks, and languages) which historically required unique implementations to utilize them. We then use this peak performance as a baseline to compare increasingly complex chains of tasks (with increasingly complex task dependencies) and evaluate how IRIS copes. Finally, we consider the performance of different IRIS scheduling policies on this range of task graphs.

Johnston, Beau↗
DOE OSTI2021

A High Performance Sparse Tensor Algebra Compiler in MLIR

Sparse tensor algebra is widely used in many applications, including scientific computing, machine learning, and data analytics. The performance of sparse tensor algebra kernels strongly depends on the intrinsic characteristics of the input tensors, hence many storage formats are designed for tensors to achieve optimal performance for particular applications/architectures, which makes it challenging to implement and optimize every tensor operation of interest on a given architecture. We propose a tensor algebra domain-specific language (DSL) and compiler framework to automatically generate kernels for mixed sparse-dense tensor algebra operations. The proposed DSL provides high-level programming abstractions that resemble the familiar Einstein notation to represent tensor algebra operations. The compiler introduces a new Sparse Tensor Algebra dialect built on top of LLVM's extensible MLIR compiler infrastructure for efficient code generation while covering a wide range of tensor storage formats. Our compiler also leverages input-dependent code optimization to enhance data locality for better performance. Our results show that the performance of automatically generated kernels outperforms the state-of-the-art sparse tensor algebra compiler, with up to 20.92x, 6.39x, and 13.9x performance improvement over state-of-the-art tensor algebra compilers, for parallel SpMV, SpMM, and TTM, respectively.

Tian, Ruiqin↗
DOE OSTI2024

XaaS: Acceleration as a Service to Enable Productive High-Performance Cloud Computing

High-performance computing (HPC) and the cloud have evolved independently, specializing their innovations into performance or productivity. Acceleration as a Service (XaaS) is a recipe to empower both fields with a shared execution platform that provides transparent access to computing resources, regardless of the underlying cloud or HPC service provider. Bridging HPC and cloud advancements, XaaS presents a unified architecture built on performance-portable containers. Here, our converged model concentrates on low-overhead, high-performance communication and computing, targeting resource-intensive workloads from climate simulations to machine learning. XaaS lifts the restricted allocation model of Function as a Service (FaaS), allowing users to benefit from the flexibility and efficient resource utilization of serverless computing while supporting long-running and performance-sensitive workloads from HPC.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Comparison of Model Predictions and Performance Test Data for a Prototype Thermal Energy Storage Module

Although model predictions of thermal energy storage (TES) performance have been explored in previous investigations, relevant test data that enable experimental validation of performance models have been limited. This is particularly true for high-performance TES designs that facilitate fast input and extraction of energy. In this paper, we present a summary of experimental tests of a high-performance TES unit using lithium nitrate trihydrate phase change material as a storage medium. Performance data are presented for complete dual-mode cycles consisting of extraction (melting) followed by charging (freezing). These tests simulate the cyclic operation of a TES unit for asynchronous cooling in a variety of applications. Finally, the model analysis is found to agree reasonably well, within 10%, with the experimental data except for conditions very near the initiation of freezing, a consequence of subcooling that is required to initiate solidification.

25 ENERGY STORAGE↗
DOE OSTI2020

SENSITIVITY ANALYSIS OF FUEL PHYSICAL PROPERTY EFFECTS ON SPARK IGNITION ENGINE PERFORMANCE

Alternative fuels are of interest to automakers and regulators due to their potential to reduce net greenhouse gas emissions from transportation sources. Alternative fuels also have fuel properties which may enable advanced combustion modes with higher engine thermal efficiencies. There has been previous work to identify the relationship between various fuel properties and engine performance, but most of this work has been experiments or simulations where the change in properties was obtained through changing the fuel composition, making isolating the effects of individual fuel properties difficult. In this study, numerical simulations have been used to investigate the effects of individual fuel physical properties such as viscosity or heat of vaporization (HoV) on engine performance. Simulations have been performed of two different engine platforms, the first an optical, single-cylinder research engine and the second a multi-cylinder production engine. Both engines are direct-injection spark-ignition engines with pent-roof heads and are designed for automotive applications. Each engine was run at a different operating condition, one stable and one knock-limited. Different base fuels provided a variety of simulated conditions. Up to six different fuel properties were varied as part of Global Sensitivity Analyses performed for each of the engines with multiple performance targets including thermal efficiency, combustion efficiency and combustion phasing. Results show trends that are largely consistent with previous experimental findings using multiple fuels. The engine thermal efficiency was primarily sensitive to the fuel's HoV, with other fuel physical properties having smaller effects. For optical engine results, the magnitude of the effect was greater in this study than expected based on previous experimental results were many fuel physical and chemical properties were varied simultaneously. However, for the multi-cylinder production engine, the relationship between thermal efficiency and HoV was slightly smaller.

engine simulation↗
DOE OSTI2024

Doping engineering: Next step toward room temperature performance of terahertz quantum cascade lasers

We hereby offer a comprehensive analysis of various factors that could potentially enable terahertz quantum cascade lasers (THz QCLs) to achieve room temperature performance. We thoroughly examine and integrate the latest findings from recent studies in the field. Our work goes beyond a mere analysis; it represents a nuanced and comprehensive exploration of the intricate factors influencing the performance of THz QCLs. Through a comprehensive and holistic approach, we propose novel insights that significantly contribute to advancing strategies for improving the temperature performance of THz QCLs. This all-encompassing perspective allows us not only to present a synthesis of existing knowledge but also to offer a fresh and nuanced strategy to improve the temperature performance of THz QCLs. We draw new conclusions from prior works, demonstrating that the key to enhancing THz QCL temperature performance involves not only optimizing interface quality but also strategically managing doping density, its spatial distribution, and profile. This is based on our results from different structures, such as two experimentally demonstrated devices: the spit-well resonant-phonon and the two-well injector direct-phonon schemes for THz QCLs, which allow efficient isolation of the laser levels from excited and continuum states. In these schemes, the doping profile has a setback that lessens the overlap of the doped region with the active laser states. Our work stands as a valuable resource for researchers seeking to gain a deeper understanding of the evolving landscape of THz technology. Furthermore, we present a novel strategy for future endeavors, providing an enhanced framework for continued exploration in this dynamic field. This strategy should pave the way to potentially reach higher temperatures than the latest records reached for T max of THz QCLs.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Operation and performance of the ATLAS tile calorimeter in LHC Run 2

The ATLAS tile calorimeter (TileCal) is the hadronic sampling calorimeter covering the central region of the ATLAS detector at the Large Hadron Collider (LHC). This paper gives an overview of the calorimeter’s operation and performance during the years 2015–2018 (Run 2). In this period, ATLAS collected proton–proton collision data at a centre-of-mass energy of 13 TeV and the TileCal was 99.65% efficient for data-taking. The signal reconstruction, the calibration procedures, and the detector operational status are presented. The performance of two ATLAS trigger systems making use of TileCal information, the minimum-bias trigger scintillators and the tile muon trigger, is discussed. Studies of radiation effects allow the degradation of the output signals at the end of the LHC and HL-LHC operations to be estimated. Finally, the TileCal response to isolated muons, hadrons and jets from proton–proton collisions is presented. The energy and time calibration methods performed excellently, resulting in good stability and uniformity of the calorimeter response during Run 2. The setting of the energy scale was performed with an uncertainty of 2%. The results demonstrate that the performance is in accordance with specifications defined in the Technical Design Report.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Chimbuko: A Workflow-Level Scalable Performance Trace Analysis Tool

Due to the sheer volume of data it is typically impractical to analyze the detailed performance of an HPC application running at-scale. While conventional small-scale benchmarking and scaling studies are often sufficient for simple applications, many modern workflow-based applications couple multiple elements with competing resource demands and complex inter-communication patterns for which performance cannot easily be studied in isolation and at small scale. This work discusses Chimbuko, a performance analysis framework that provides real-time, in situ anomaly detection. By focusing specifically on performance anomalies and their origin (aka provenance), data volumes are dramatically reduced without losing necessary details. To the best of our knowledge, Chimbuko is the first online, distributed, and scalable workflow-level performance trace analysis framework. We demonstrate the tool's usefulness on Oak Ridge National Laboratory's Summit system.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

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↗
DOE OSTI2024

Performance Study of CXL Memory Topology

This paper presents a comprehensive evaluation of the performance impact of various Compute Express Link (CXL) memory topologies, with a particular emphasis on CXL switches, in the context of High- Performance Computing (HPC) and Large Language Model (LLM) inference workloads. Our study unveils significant performance variations across different topologies, demonstrating that certain configurations yield superior performance for specific workloads. These findings underscore the critical importance of tailored topol- ogy selection in optimizing system performance. Additionally, we address the inherent challenges associated with integrating CXL switches, including overhead considerations and routing complex- ities. Our research highlights the necessity for thorough evalua- tion methodologies to fully leverage CXL technology’s potential in contemporary computing environments. These insights provide valuable guidance for system architects and data center operators in designing and optimizing CXL-based infrastructures for diverse workload requirements.

CXL, memory, Artificial Intelligence (AI), HPC↗
DOE OSTI2025

HP-MDR: High-performance and Portable Data Refactoring and Progressive Retrieval with Advanced GPUs

Scientific applications produce vast amounts of data, posing grand challenges in the underlying data management and analytic tasks. Progressive compression is a promising way to address this problem, as it allows for on-demand data retrieval with significantly reduced data movement cost. However, most existing progressive methods are designed for CPUs, leaving a gap for them to unleash the power of today’s heterogeneous computing systems with GPUs.In this work, we propose HP-MDR, a high-performance and portable data refactoring and progressive retrieval framework for GPUs. Our contributions are four-fold: (1) We carefully optimize the bitplane encoding and lossless encoding, two key stages in progressive methods, to achieve high performance on GPUs; (2) We propose pipeline optimization and incorporate it with data refactoring and progressive retrieval workflows to further enhance the performance for large data process; (3) We leverage our framework to enable high-performance data retrieval with guaranteed error control for common Quantities of Interest; (4) We evaluate HP-MDR and compare it with state of the arts using five real-world datasets. Experimental results demonstrate that HP-MDR delivers an average 13.68 × and 6.31 × throughput in data refactoring and progressive retrieval tasks, respectively. It also leads to 11.22 × throughput for recomposing required data representations under Quantity-of-Interest error control and 6.04 × performance for the corresponding end-to-end data retrieval, when compared with state-of-the-art solutions.

Li, Yanliang [University of Oregon]↗
DOE OSTI2025

A Study of Performance Portability of Low-bit Fused Matrix-Vector Multiplication Kernels in SYCL

Understanding the causes of performance gaps between a portable programming model and a vendor-specific programming model is important for improving performance portability. This paper studies performance portability of low-bit fused general matrix-vector multiplication kernels in SYCL on vendors’ graphics processing units (GPUs). This work introduces the use case, explains the kernel implementations in detail, evaluates the performance of the CUDA, HIP, and SYCL kernels on datacenter, desktop, and laptop GPUs, and investigates the causes of performance gaps. The results show that loop unrolling, kernel dispatch overhead, and sum reduction contribute to the gaps.

Jin, Zheming [ORNL] (ORCID:000000027197780X)↗
DOE OSTI2021

Effects of Ink Formulation on the Structure and Performance of PGM-Free Catalyst Layer in PEMFCs

Platinum group metal (PGM) catalysts are the major electrocatalysts for oxygen reduction reaction (ORR) in the polymer electrolyte membrane fuel cells (PEMFCs). The cost becomes unaffordable if the PEMFC is in massive application. The PGM-Free catalyst shows very promising activity in rotation disk electrode (RDE) testing. The half-wave potential could reach 0.91 V versus standard hydrogen electrode (SHE). However, in a membrane electrode assembly (MEA), the performance of PGM-Free catalysts is not good enough to replace the PGM catalysts. Since the PGM-free catalysts are so different from the PGM catalysts in terms of catalytic activity, stability, surface conditions, particle size, etc., the fabrication of PGM-Free catalyst MEA cannot simply copy the method of making PGM MEA. Here we proposed a novel method of fabricating PGM-Free catalyst MEA, so that the intrinsic catalyst activity from RDE can be translated into MEA performance. The method is based on the catalyst coated membrane (CCM) method using optimized ionomer to carbon (I/C) ratio and solvent mixture of catalyst ink. Using this method, the PGM-free catalyst MEA achieved the current density 44.9 mA cm -2 at 0.9 V iR-free in H 2 /O 2 and 150 mA cm -2 at 0.8 V in H 2 /air, which surpassed the performance targets of US Department of Energy (DOE)for PGM-Free catalyst MEA. The property (solvent composition, dispersion of catalyst and ionomer in an ink), structure (pore structure) and the MEA performance have been characterized using, mercury intrusion porosimetry (MIP), MEA testing. A property-structure-performance relationship has been established.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Impact of Pressure Distribution and Magnitude on the Performance of Lithium Metal Anodes

Li metal anodes are a critical battery technology due to their ability to substantially increase the energy density of Li-based batteries. It is well known that pressure greatly impacts the performance of a Li-metal anode. However, precisely how the pressure value and distribution of pressure affect performance is unclear. Furthermore, the solid-electrolyte interphase composition that forms under varying pressure distributions remains a key parameter for practical lithium metal anodes. In this work, different pressure distributions were employed by using differently shaped and oriented mechanical springs in the coin cells, resulting in varying contact points. Pressure-sensitive films were used to spatially map the pressure and correlate it to the performance. It was found that higher average pressure does not necessarily have a positive effect on performance. When high pressure is paired with poor pressure uniformity, the performance is in fact worse likely due to the current focusing effect, rendering unsatisfied cycling stability. This work points to the importance of controlling the relationship between average pressure and pressure uniformity.

25 ENERGY STORAGE↗
DOE OSTI2024

Performance Losses and Current-Driven Recovery from Cation Contaminants in PEM Water Electrolysis

Water contaminants are a common cause of failure for polymer electrolyte membrane (PEM) electrolyzers in the field as well as a confounding factor in research on cell performance and durability. In this study, we investigated the performance impacts of feed water containing representative tap water cations at concentrations ranging from 0.5–500 μ M, with conductivities spanning from ASTM Type II to tap-water levels. We present multiple diagnostic signatures to help identify the presence of contaminants in PEM electrolysis cells. Through analysis of polarization curves and impedance spectroscopy to understand the origins of performance losses, we found that a switch from the acidic to alkaline hydrogen evolution mechanism is a key factor in contaminated cell behavior. Finally, we demonstrated that this mechanism switching can be harnessed to remove cation contaminants and recover cell performance without the use of an acid wash. We demonstrated near-complete recovery of cells contaminated with sodium and calcium, and partial recovery of a cell contaminated with iron, which was further investigated by post-mortem microscopy. The improved understanding of contaminant impacts from this work can inform development of strategies to mitigate or recover performance losses as well as improve the consistency and rigor of electrolysis research.

30 DIRECT ENERGY CONVERSION↗