Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Computer architecture”

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 1,747 records · Page 97

Real-Time Lifetime Prediction of Semiconductor Devices Using Hardware-in-the-Loop

This paper presents a unique approach to enable real-time lifespan prediction of semiconductor power modules using a Hardware-in-the-Loop (HIL) system. By integrating the module's overall loss characteristics-specifically switching and conduction losses-with a thermoelectric model of the thermal management system, this research demonstrates that the model can dynamically estimates the junction temperature profile of the semiconductor devices in response to a changing torque demand profile for the motor drive system. This capability enables continuous monitoring of the module's operational time and cumulative stress induced on the devices to compute accumulated remaining lifetime or time-to-failure (TTF). This study provides an architectural framework for the HIL system with high-fidelity component models of multiple physical domains, allowing simulation of dynamic behaviors of a closely-coupled motor drive system. The advanced real-time computation and measurement functionalities of the HIL system allow for both dynamic lifetime calculations based on simulated data and aggregate lifetime predictions utilizing historical data. Moreover, this paper details an algorithm that not only computes cumulative damage but also synthesizes these data into a comprehensive aggregated lifetime metric. This methodology can enhance the maintenance scheduling strategies and operational reliability of semiconductor devices in critical applications, ultimately extending their service life while optimizing performance.

hardware-in-the-loop (HIL)↗

A conserved RNA structural motif for organizing topology within picornaviral internal ribosome entry sites

Picornaviral IRES elements are essential for initiating the cap-independent viral translation. However, three-dimensional structures of these elements remain elusive. Here, we report a 2.84-Å resolution crystal structure of hepatitis A virus IRES domain V (dV) in complex with a synthetic antibody fragment—a crystallization chaperone. The RNA adopts a three-way junction structure, topologically organized by an adenine-rich stem-loop motif. Despite no obvious sequence homology, the dV architecture shows a striking similarity to a circularly permuted form of encephalomyocarditis virus J-K domain, suggesting a conserved strategy for organizing the domain architecture. Recurrence of the motif led us to use homology modeling tools to compute a 3-dimensional structure of the corresponding domain of foot-and-mouth disease virus, revealing an analogous domain organizing motif. The topological conservation observed among these IRESs and other viral domains implicates a structured three-way junction as an architectural scaffold to pre-organize helical domains for recruiting the translation initiation machinery.

59 BASIC BIOLOGICAL SCIENCES↗

Fast and Accurate Simulation of the Cray XMT Multithreaded Supercomputer

Irregular applications, such as data mining and analysis or graph-based computations, show unpredictable memory/network access patterns and control structures. Highly multithreaded architectures with large processor counts, like the Cray MTA-1, MTA-2 and XMT, appear to address their requirements better than commodity clusters. However, the research on highly multithreaded systems is currently limited by the lack of adequate architectural simulation infrastructures due to issues such as size of the machines, memory footprint, simulation speed, accuracy and customization. At the same time, Shared-memory MultiProcessors (SMPs) with multi-core processors have become an attractive platform to simulate large scale machines. In this paper, we introduce a cycle-level simulator of the highly multithreaded Cray XMT supercomputer. The simulator runs unmodified XMT applications. We discuss how we tackled the challenges posed by its development, detailing the techniques introduced to make the simulation as fast as possible while maintaining a high accuracy. By mapping XMT processors (ThreadStorm with 128 hardware threads) to host computing cores, the simulation speed remains constant as the number of simulated processors increases, up to the number of available host cores. The simulator supports zero-overhead switching among different accuracy levels at run-time and includes a network model that takes into account contention. On a modern 48-core SMP host, our infrastructure simulates a large set of irregular applications 500 to 2000 times slower than real time when compared to a 128-processor XMT, while remaining within 10\% of accuracy. Emulation is only from 25 to 200 times slower than real time.

Cray XMT↗

Extending SEER for Extreme Heterogeneity

Heterogeneous and multi-device nodes are increasingly common in high-performance computing and data centers, yet existing programming models often lack simple, transparent, and portable support for these diverse architectures. The main contribution of this work is the development of novel SEER capabilities to address this challenge by providing a descriptive programming model that allows applications to seamlessly leverage heterogeneous nodes across various device types. SEER uses efficient memory management and can select the proper device[s] depending on the computational cost of the applications. This is completely transparent to the programmer, thereby providing a highly productive programming environment. Integrating extreme heterogeneity into the SEER library as shown with the use of NVIDIA and AMD GPUs simultaneously allows it to expand and exploit the performance possibilities. Our analysis based on the well-known Conjugate Gradient algorithm reports accelerations above 1.5 × on computationally demanding steps of such an algorithm by using both architectures simultaneously.

Teranishi, Keita [ORNL] (ORCID:0000000166472690)↗

Gyrokinetic particle-in-cell optimization on emerging multi- and manycore platforms

The next decade of high-performance computing (HPC) systems will see a rapid evolution and divergence of multi- and manycore architectures as power and cooling constraints limit increases in microprocessor clock speeds. Understanding efficient optimization methodologies on diverse multicore designs in the context of demanding numerical methods is one of the greatest challenges faced today by the HPC community. In this paper, we examine the efficient multicore optimization of GTC, a petascale gyrokinetic toroidal fusion code for studying plasma microturbulence in tokamak devices. For GTC’s key computational components (charge deposition and particle push), we explore efficient parallelization strategies across a broad range of emerging multicore designs, including the recently-released Intel Nehalem-EX, the AMD Opteron Istanbul, and the highly multithreaded Sun UltraSparc T2+. We also present the first study on tuning gyrokinetic particle-in-cell (PIC) algorithms for graphics processors, using the NVIDIA C2050 (Fermi). Our work discusses several novel optimization approaches for gyrokinetic PIC, including mixed-precision computation, particle binning and decomposition strategies, grid replication, SIMDized atomic floating-point operations, and effective GPU texture memory utilization. Overall, we achieve significant performance improvements of 1.3–4.7× on these complex PIC kernels, despite the inherent challenges of data dependency and locality. Finally, our work also points to several architectural and programming features that could significantly enhance PIC performance and productivity on next-generation architectures.

97 MATHEMATICS AND COMPUTING↗

Ionothermally Synthesized Nanoporous Ti 0.95 W 0.05 Nb 2 O 7 : a Novel Anode Material for High-Performance Lithium-Ion Batteries

Although TiNb 2 O 7 is regarded as a fast-rechargeable lithium-ion battery (LIB) anode material, the intrinsic poor electrochemical kinetics of TiNb 2 O 7 still dramatically impedes its development. Herein, an ionothermal synthesis-assisted doping strategy is proposed for the preparation of a new W 6+ -doped TiNb 2 O 7 material (Ti 0.95 W 0.05 Nb 2 O 7 ) with nanoporous structure (denoted as NPTWNO). The improved Li + diffusion coefficient of NPTWNO suggests that the ionic-liquid-templated nanoporous architecture improves the Li+ diffusion kinetics. The density functional theory computational study reveals that the doped W 6+ successfully boosts the electronic conductivity due to the narrowed conduction-valance bandgap resulted from charge redistribution, which is reflected by the electrochemical impedance spectroscopy data. With the simultaneously enhanced Li+ diffusivity and electronic conductivity, NPTWNO achieves fast-rechargeability in LIBs. So this work indicates the potential of ionothermal synthesis-assisted doping strategy on energy storage materials and offers NPTWNO material with promising electrochemical performance.

25 ENERGY STORAGE↗

The Archive Solution for Distributed Workflow Management Agents of the CMS Experiment at LHC

The CMS experiment at the CERN LHC developed the Workflow Management Archive system to persistently store unstructured framework job report documents produced by distributed workflow management agents. In this paper we present its architecture, implementation, deployment, and integration with the CMS and CERN computing infrastructures, such as central HDFS and Hadoop Spark cluster. The system leverages modern technologies such as a document oriented database and the Hadoop eco-system to provide the necessary flexibility to reliably process, store, and aggregate $\mathcal{O}$(1M) documents on a daily basis. We describe the data transformation, the short and long term storage layers, the query language, along with the aggregation pipeline developed to visualize various performance metrics to assist CMS data operators in assessing the performance of the CMS computing system.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

How the Size and Density of Charge-Transfer Excitons Depend on Heterojunction’s Architecture

We have characterized the size, intensity, density, and distribution of charge-transfer (CT) excitons as a function of the acceptor–donor architecture of prototypical organic interfaces. This characterization was done by computational analysis of 17 models of varying numbers, positions, and orientations of the donor and acceptor molecules. The models’ building blocks were phenyl-C61-butyric acid methyl ester (PCBM) fullerene acceptors and dual-band donor polymers composed of thiophene, benzothiadiazole, and benzotriazole subunits. The electronic structure of the donor–acceptor complexes was computed with the time-dependent long-range-corrected density-functional tight-binding method and analyzed with the fragment-based one-electron transition density matrix. In all models, the complexes with edge-on orientation have denser spectra of low-energy CT states lying below the absorption bands compared to the complexes with face-on orientation. This CT-state distribution in edge-on complexes provides a gate to efficiently populate cold CT excitons. Moreover, the cold CT excitons have a higher degree of charge separation in the edge-on than in the face-on complexes. The CT amount and the CT exciton size generally increase with the energy of the CT states, although the electron remains localized on a single molecule in cold CT states. Delocalization over two PCBM molecules was observed for high-energy CT states. The exciton size also depends on the orientation. Larger excitons are produced by the delocalization of the electrons perpendicularly to the interface. When the delocalization is parallel, the smaller electron–hole distances yield moderately sized CT excitons.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Dangling Bonds as Possible Contributors to Charge Noise in Silicon and Silicon–Germanium Quantum Dot Qubits

Spin qubits based on Si and Si 1–x Ge x quantum dot architectures exhibit among the best coherence times of competing quantum computing technologies, yet they still suffer from charge noise that limit their qubit gate fidelities. Identifying the origins of these charge fluctuations is therefore a critical step toward improving Si quantum-dot-based qubits. Here, we use hybrid functional calculations to investigate possible atomistic sources of charge noise, focusing on charge trapping at Si and Ge dangling bonds (DBs). We evaluate the role of global and local environment in the defect levels associated with DBs in Si, Ge, and Si 1–x Ge x alloys, and consider their trapping and excitation energies within the framework of configuration coordinate diagrams. We additionally consider the influence of strain and oxidation in charge-trapping energetics by analyzing Si and GeSi DBs in SiO 2 and strained Si layers in typical Si 1–x Ge x quantum dot heterostructures. Our results identify that Ge dangling bonds are more problematic charge-trapping centers both in typical Si 1–x Ge x alloys and associated oxidation layers, and they may be exacerbated by compositional inhomogeneities. In conclusion, these results suggest the importance of alloy homogeneity and possible passivation schemes for DBs in Si-based quantum dot qubits and are of general relevance to mitigating possible trap levels in other Si, Ge, and Si 1–x Ge x -based metal-oxide-semiconductor stacks and related devices.

36 MATERIALS SCIENCE↗

Fourier-based three-dimensional multistage transformer for aberration correction in multicellular specimens

High-resolution tissue imaging is often compromised by sample-induced optical aberrations that degrade resolution and contrast. Although wavefront sensor-based adaptive optics (AO) can measure these aberrations, such hardware solutions are typically complex, expensive to implement and slow when serially mapping spatially varying aberrations across large fields of view. Here we introduce AOViFT (adaptive optical vision Fourier transformer)—a machine learning-based aberration sensing framework built around a three-dimensional multistage vision transformer that operates on Fourier domain embeddings. AOViFT infers aberrations and restores diffraction-limited performance in puncta-labeled specimens with substantially reduced computational cost, training time and memory footprint compared to conventional architectures or real-space networks. We validated AOViFT on live gene-edited zebrafish embryos, demonstrating its ability to correct spatially varying aberrations using either a deformable mirror or postacquisition deconvolution. By eliminating the need for the guide star and wavefront sensing hardware and simplifying the experimental workflow, AOViFT lowers technical barriers for high-resolution volumetric microscopy across diverse biological samples.

Alshaabi, Thayer [Howard Hughes Medical Institute,↗

Parallel interior-point solver for block-structured nonlinear programs on SIMD/GPU architectures

Here, we investigate how to port the standard interior-point method to new exascale architectures for block-structured nonlinear programs with state equations. Computationally, we decompose the interior-point algorithm into two successive operations: the evaluation of the derivatives and the solution of the associated Karush-Kuhn-Tucker (KKT) linear system. Our method accelerates both operations using two levels of parallelism. First, we distribute the computations on multiple processes using coarse parallelism. Second, each process uses SIMD/GPU accelerators locally to accelerate the operations using fine-grained parallelism. The KKT system is reduced by eliminating the inequalities and the state variables from the corresponding equations. We demonstrate our method's capability on the supercomputer Polaris, a testbed for the future exascale Aurora system. Each node is equipped with four GPUs, a setup amenable to our two-level approach. Our experiments on the stochastic optimal power flow problem show that the reduction method is 50x faster than the sparse linear solver HSL MA57 running in serial on the CPU, and 6x faster than Pardiso running in parallel on CPU on the same number of processes.

97 MATHEMATICS AND COMPUTING↗

Toward Robust Autotuning of Noisy Quantum dot Devices

The current autotuning approaches for quantum dot (QD) devices, while showing some success, lack an assessment of data reliability. This leads to unexpected failures when noisy or otherwise low-quality data is processed by an autonomous system. In this work, we propose a framework for robust autotuning of QD devices that combines a machine learning (ML) state classifier with a data quality control module. The data quality control module acts as a “gatekeeper” system, ensuring that only reliable data are processed by the state classifier. Lower data quality results in either device recalibration or termination. To train both ML systems, we enhance the QD simulation by incorporating synthetic noise typical of QD experiments. We confirm that the inclusion of synthetic noise in the training of the state classifier significantly improves the performance, resulting in an accuracy of 95.0(9)% when tested on experimental data. We then validate the functionality of the data quality control module by showing that the state classifier performance deteriorates with decreasing data quality, as expected. Our results establish a robust and flexible ML framework for autonomous tuning of noisy QD devices.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Strategies for Integrating Deep Learning Surrogate Models with HPC Simulation Applications

The emerging trend of the convergence of high performance computing (HPC), machine learning/deep learning (ML/DL), and big data analytics presents a host of challenges for large-scale computing campaigns that seek best practices to interleave traditional scientific simulation-based workloads with ML/DL models. A portfolio of systematic approaches to incorporate deep learning into modeling and simulation serves a vital need when we support AI for science at a computing facility. In this paper, we evaluate several strategies for deploying deep learning surrogate models in a representative physics application on supercomputers at the Oak Ridge Leadership Computing Facility (OLCF). We discuss a set of recommended deployment architectures and implementation approaches. We analyze and evaluate these alternatives and show their performance and scalability up to 1000 GPUs on two mainstream platforms equipped with different deep learning hardware and software stacks.

Yin, Junqi↗

An Alternative Timing and Synchronization Approach for Situational Awareness and Predictive Analytics

Accurate and synchronized timing information is required by power system operators for controlling the grid infrastructure (relays, Phasor Measurement Units (PMUs), etc.) and determining asset positions. Satellite-based global positioning system (GPS) is the primary source of timing information. However, GPS disruptions today (both intentional and unintentional) can significantly compromise the reliability and security of our electric grids. A robust alternate source for accurate timing is critical to serve both as a deterrent against malicious attacks and as a redundant system in enhancing the resilience against extreme events that could disrupt the GPS network. To achieve this, we rely on the highly accurate, terrestrial atomic clock-based network for alternative timing and synchronization. In this paper, we discuss an experimental setup for an alternative timing approach. The data obtained from this experimental setup is continuously monitored and analyzed using various time deviation metrics. We also use these metrics to compute deviations of our clock with respect to the National Institute of Standards and Technologys (NIST) GPS data. The results obtained from these metric computations are elaborately discussed. Finally, we discuss the integration of the procedures involved, like real-time data ingestion, metric computation, and result visualization, in a novel microservices-based architecture for situational awareness.

Chinthavali, Supriya↗

Accelerating DCA++ (Dynamical Cluster Approximation) Scientific Application on the Summit supercomputer

Optimizing scientific applications on today’s accelerator-based high performance computing systems can be challenging, especially when multiple GPUs and CPUs with heterogeneous memories and persistent non-volatile memories are present. An example is Summit, an accelerator-based system at the Oak Ridge Leadership Computing Facility (OLCF) that is rated as the world’s fastest supercomputer to-date. New strategies are thus needed to expose the parallelism in legacy applications, while being amenable to efficient mapping to the underlying architecture.In this paper we discuss our experiences and strategies to port a scientific application, DCA++, to Summit. DCA++ is a highperformance research application that solves quantum manybody problems with a cutting edge quantum cluster algorithm, the dynamical cluster approximation.Our strategies aim to synergize the strengths of the different programming models in the code. These include: (a) streamlining the interactions between the CPU threads and the GPUs, (b) implementing computing kernels on the GPUs and decreasing CPU-GPU memory transfers, (c) allowing asynchronous GPU communications, and (d) increasing compute intensity by combining linear algebraic operations.Full-scale production runs using all 4600 Summit nodes attained a peak performance of 73.5 PFLOPS with a mixed precision implementation.We observed a perfect strong and weak scaling for the quantum Monte Carlo solver in DCA++, while encountering about 2× input/output (I/O) and MPI communication overhead on the time-to-solution for the full machine run. Our hardware agnostic optimizations are designed to alleviate the communication and I/O challenges observed, while improving the compute intensity and obtaining optimal performance on a complex, hybrid architecture like Summit.

Balduzzi, Giovanni↗

Image transformers for classifying acute lymphoblastic leukemia

Cancer is the leading cause of death by disease in American children. Each year, nearly 16,000 children in the United States and over 300,000 children globally are diagnosed with cancer. Leukemia is a form of blood cancer that originates in the bone marrow and accounts for one-third of pediatric cancers. This disease occurs when the bone marrow contains 20% or more immature white blood cell blasts. Acute lymphoblastic leukemia is the most prevalent leukemia type found in children, with half of all annual cases in the U.S. diagnosed for subjects under 20 years of age. To diagnose acute lymphoblastic leukemia, pathologists often conduct a morphological bone marrow assessment. This assessment determines whether the immature white blood cell blasts in bone marrow display the correct morphological characteristics, such as size and appearance of nuclei. Pathologists also use immunophenotyping via multi-channel flow cytometry to test whether certain antigens are present on the surface of blast cells; the antigens are used to identify the cell lineage of acute lymphoblastic leukemia. These manual processes require well-trained personnel and medical professionals, thus being costly in time and expenses. Computerized decision support via machine learning can accelerate the diagnosis process and reduce the cost. Training a reliable classification model to distinguish between mature and immature white blood cells is essential to the decision support system. Here, we adopted the Vision Transformer model to classify white blood cells. The Vision Transformer achieved superb classification performance compared to state-of-the-art convolutional neural networks while requiring less computational resources for training. Additionally, the latent self-attention architecture provided attention maps for a given image, providing clues as to which portion(s) of the image were significant in decision-making. We applied the Vision Transformer model and a convolutional neural network model to an acute lymphoblastic leukemia classification dataset of 12,528 samples and achieved accuracies of 88.4% and 86.2%.

Cho, Priscilla↗

Know Your Space: Inlier and Outlier Construction for Calibrating Medical OOD Detectors

This software offers methods and functions for training calibrated out-of-distribution detectors for medical image classification tasks. It includes functionalities for training, synthesizing data augmentations, calibration, and out-of-distribution detection. Developed using PyTorch, this software is compatible with standard neural network architectures used for imaging data. Additionally, it provides capabilities to compute evaluation metrics for assessing the performance and quality of the detectors.

Narayanaswamy, VivekSivaraman↗