Engineering topics

Wang, Xiao

Publications and source records attributed to Wang, Xiao.

11 records

DOE OSTI2025

Ruthenium/Carbon Nanocomposites for Efficient Hydrogen Electrocatalysis: Impacts of Halide Residues

Ruthenium has emerged as a promising substitute for platinum toward the hydrogen evolution/oxidation reaction (HER/HOR). Herein, ruthenium/carbon composites are prepared by magnetic induction heating (300 A, 10 s) of RuCl3, RuBr3 or RuI3 loaded on hollow N-doped carbon cages (HNC). The HNC-RuCl3-300A sample consists of Ru nanoparticles (dia. 1.96 nm) and abundant Cl residues. HNC-RuBr3-300A possesses a larger nanoparticle size (≈19.36 nm) and lower content of Br residues. HNC-RuI3-300A contains only bulk-like Ru agglomerates with a minimal amount of I residues, due to reduced Ru-halide bonding interactions. Among these, HNC-RuCl3-300A exhibits the best HER activity in alkaline media, with a low overpotential of only -26 mV to reach 10 mA cm-2, even outperforming Pt/C, and can be used as the cathode catalyst for anion exchange membrane water electrolyzer (along with commercial RuO2 as the anode catalyst), producing 0.5 A cm- 2 at 1.88 V for up to 100 h, a performance markedly better than that with Pt/C. HNC-RuCl3-300A also exhibits the best HOR activity, with a half-wave potential (+18 mV) even lower than that of Pt/C (+35 mV). These activities are ascribed to the combined contributions of small Ru nanoparticles and Ru-to-halide charge transfer that weaken H adsorption.

Yu, Bingzhe↗

DOE OSTI2025

Mechanistic studies of small molecule ligands selective to RNA single G bulges

Abstract Small-molecule RNA binders have emerged as an important pharmacological modality. A profound understanding of the ligand selectivity, binding mode, and influential factors governing ligand engagement with RNA targets is the foundation for rational ligand design. Here, we report a novel class of coumarin derivatives exhibiting selective binding affinity towards single G RNA bulges. Harnessing the computational power of all-atom Gaussian accelerated molecular dynamics simulations, we unveiled a rare minor groove binding mode of the ligand with a key interaction between the coumarin moiety and the G bulge. This predicted binding mode is consistent with results obtained from structure-activity relationship studies and transverse relaxation measurements by nuclear magnetic resonance spectroscopy. We further generated 444 molecular descriptors from 69 coumarin derivatives and identified key contributors to the binding events, such as charge state and planarity, by lasso (least absolute shrinkage and selection operator) regression. Our work deepened the understanding of RNA-small molecule interactions and integrated a new framework for the rational design of selective small-molecule RNA binders.

Biochemistry & Molecular Biology↗

DOE OSTI2025

Modulation of lignin and anthocyanin homeostasis by GTP cyclohydrolase1 in maize

Summary Maize is a key biomass resource with wide agricultural applications. Anthocyanins, potent antioxidants, offer health benefits like reducing oxidative stress. The biosynthesis of anthocyanins competes with that of lignin for shared metabolic precursors, which can lead to trade‐offs in plant growth and feed quality. Higher lignin content can decrease silage digestibility, posing challenges for livestock feed. The maizebrown midrib 6(bm6) mutant, known for reduced lignin, has an unclear genetic basis. Here, we identifyZmGCH1as the candidate gene forbm6through fine mapping. Mutations inZmGCH1shift precursors from lignin to anthocyanin biosynthesis. Furthermore, we show that ZmGCH1 interacts with ZmPEBP15 to modulate chalcone synthase activity, thereby stabilizing the allocation of precursors between lignin and anthocyanin pathways. To evaluate the practical implications of our findings, we introduced thebm6mutation into Zhengdan958 and Xianyu335. In vitro rumen digestion assays confirmed that the introduction of thebm6mutation significantly improved silage digestibility. This discovery not only holds great potential for enhancing silage digestibility but also provides a broader strategy for optimizing maize production to better meet the increasing demands of both the food and livestock feed.

Biotechnology & Applied Microbiology↗

DOE OSTI2024

ORBIT: Oak Ridge Base Foundation Model for Earth System Predictability

Earth system predictability is challenged by the complexity of environmental dynamics and the multitude of variables involved. Current AI foundation models, although advanced by leveraging large and heterogeneous data, are often constrained by their size and data integration, limiting their effectiveness in addressing the full range of Earth system prediction challenges. To overcome these limitations, we introduce the Oak Ridge Base Foundation Model for Earth System Predictability (ORBIT), an advanced vision transformer model that scales up to 113 billion parameters using a novel hybrid tensor-data orthogonal parallelism technique. As the largest model of its kind, ORBIT surpasses the current climate AI foundation model size by a thousandfold. Performance scaling tests conducted on the Frontier supercomputer have demonstrated that ORBIT achieves 684 petaFLOPS to 1.6 exaFLOPS sustained throughput, with scaling efficiency maintained at 41% to 85% across 49,152 AMD GPUs. These breakthroughs establish new advances in AI-driven climate modeling and demonstrate promise to significantly improve the Earth system predictability.

Wang, Xiao↗

DOE OSTI2024

autoGEMM: Pushing the Limits of Irregular Matrix Multiplication on Arm Architectures

This paper presents an open-source library that pushes the limits of performance portability for irregular General Matrix Multiplication (GEMM) on the widely-used Arm architectures. Our library, autoGEMM, is designed to support a wide range of Arm processors: from edge devices to HPC-grade CPUs. autoGEMM generates optimized kernels for various hardware configurations by auto-combining fragments of autogenerated micro-kernels that employ hand-written optimizations to maximize computational efficiency. We optimize the kernel pipeline by tuning the register reuse and the data load/store overlapping. In addition, we use a dynamic tiling scheme to generate balanced tile shapes. Finally, we position autoGEMM on top of the TVM framework where our dynamic tiling scheme prunes the search space for TVM to identify the optimal combination of parameters for code optimization. Evaluations on five different classes of Arm chips demonstrate the advantages of autoGEMM. For small matrices, autoGEMM achieves 98% of peak and up to 2.0x speedup over state-of-the-art libraries such as LIBXSMM and LibShalom. For irregular matrices (i.e. tall skinny and long rectangles), autoGEMM is 1.3-2.0x faster than widely-used libraries such as OpenBLAS and Eigen. autoGEMM is publicly available at: https://github.com/wudu98/autoGEMM.

Wu, Du↗

DOE OSTI2024

Adaptive Patching for High-resolution Image Segmentation with Transformers

Attention-based models are proliferating in the space of image analytics, including segmentation. The standard method of feeding images to transformer encoders is to divide the images into patches and then feed the patches to the model as a linear sequence of tokens. For high-resolution images, e.g. microscopic pathology images, the quadratic compute and memory cost prohibits the use of an attention-based model, if we are to use smaller patch sizes that are favorable in segmentation. The solution is to either use custom complex multi-resolution models or approximate attention schemes. We take inspiration from Adapative Mesh Refinement (AMR) methods in HPC by adaptively patching the images, as a pre-processing step, based on the image details to reduce the number of patches being fed to the model, by orders of magnitude. This method has a negligible overhead, and works seamlessly with any attention-based model, i.e. it is a pre-processing step that can be adopted by any attention-based model without friction. We demonstrate superior segmentation quality over SoTA segmentation models for realworld pathology datasets while gaining a geomean speedup of 6.9× for resolutions up to 64K2, on up to 2, 048 GPUs.

Zhang, Enzhi↗

DOE OSTI2024

Neural architecture search via similarity adaptive guidance

Evolutionary neural network architecture search (ENAS) has attracted the attention of many experts due to its global optimization capabilities to automatically search for convolutional neural network architectures based on the target task. The current search space for ENAS is not to design a fully structured network, but to search for smaller cell architectures to reduce search costs. However, blind search strategies do not effectively utilize the potential experience of the population. In order to utilize the potential experience learned by the current population to guide the evolutionary search of the population, we propose a similarity guided neural network architecture search algorithm based on cell architecture, which utilizes the similarity between pairwise architectures in the population as empirical knowledge learned by the population. Our proposed algorithm provides a novel method for calculating architecture similarity, which calculates architecture similarity separately from the cell and macro-structure. Then we decouple the connections and operations in the cell and calculate connection and operation similarity separately. In addition, we propose adaptive similarity selection and binary tournament selection strategies to enhance the algorithm’s global and local search capabilities and effectively explore the search space. Finally, we design an improved single-point crossover operator to enhance the local search ability of the evolutionary operator. The experimental results show that SAGNAS is a competitive algorithm that achieves 97.44% and 81.60% in CIFAR10 and CIFAR100 with only 1.9 GPU-days spent.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Real-time High-resolution X-Ray Computed Tomography

Computed Tomography (CT) serves as a key imaging technology that relies on computationally intensive filtering and back-projection algorithms for 3D image reconstruction. While conventional high-resolution image reconstruction (> 2K3) solutions provide quick results, they typically treat reconstruction as an offline workload to be performed remotely on large-scale HPC systems. The growing demand for post-construction AI-driven analytics and the need for real-time adjustments call for high-resolution reconstruction solutions that are feasible on local computing resources, i.e. a multi-GPU server at most. In this paper, we propose a novel approach that utilizes Tensor Cores to optimize image reconstruction without sacrificing precision. We also introduce a framework designed to enable real-time execution of end-to-end distributed image reconstruction in a multi-GPU environment. Evaluations conducted on a single Nvidia A100 and H100 GPU show performance improvements of 1.91 × and 2.15 × compared to highly optimized production libraries. Furthermore, our framework, when deployed on 8-card Nvidia A100 GPU system, demonstrates the ability to reconstruct real-world datasets into 20483 volumes (32 GB) in slightly more than one minute and 40963 volumes (256 GB) in 7 minutes.

Wu, Du↗

DOE OSTI2024

Integrated photonic encoder for low power and high-speed image processing

Abstract Modern lens designs are capable of resolving greater than 10 gigapixels, while advances in camera frame-rate and hyperspectral imaging have made data acquisition rates of Terapixel/second a real possibility. The main bottlenecks preventing such high data-rate systems are power consumption and data storage. In this work, we show that analog photonic encoders could address this challenge, enabling high-speed image compression using orders-of-magnitude lower power than digital electronics. Our approach relies on a silicon-photonics front-end to compress raw image data, foregoing energy-intensive image conditioning and reducing data storage requirements. The compression scheme uses a passive disordered photonic structure to perform kernel-type random projections of the raw image data with minimal power consumption and low latency. A back-end neural network can then reconstruct the original images with structural similarity exceeding 90%. This scheme has the potential to process data streams exceeding Terapixel/second using less than 100 fJ/pixel, providing a path to ultra-high-resolution data and image acquisition systems.

47 OTHER INSTRUMENTATION↗

NASA NTRS2005

Characterization and recovery of Deep Sub Micron (DSM) technologies behavior under radiation

This paper serves a twofold purpose: characterize the behavior of a reconfigurable chip exposed to radiation; and demonstrate a method for functionality recovery due to Total Ionizing Dose (TID) effects. The experiments are performed using a PL developed reconfigurable device, a Field Programmable Transistor Array (FPTA). The paper initially describes experiments on the characterization of the NMOS transistor behavior for TID values up to 300krad. The behavior of analog and digital circuits downloaded onto the FPTA chip is also assessed for TID effects. This paper also presents a novel approach for circuit functionality recovery due to radiation effects based on Evolvable Hardware. The key idea is to reconfigure a programmable device, in-situ, to compensate, or bypass its degraded or damaged components. Experiments with total radiation dose up to 300kRad show that while the functionality of a variety of circuits, including digital gates, a rectifier and a Digital to Analog Converter implemented on a FPTA-2 chip is degraded/lost at levels before 200kRad, the correct functionality can be recovered through the proposed evolutionary approach and the chips are able to survive higher radiation, for several functions in excess of total radiation dose of 250kRad.

evolvable hardware↗

NASA NTRS2002

Tissue engineering skeletal muscle for orthopaedic applications

With current technology, tissue-engineered skeletal muscle analogues (bioartificial muscles) generate too little active force to be clinically useful in orthopaedic applications. They have been engineered genetically with numerous transgenes (growth hormone, insulinlike growth factor-1, erythropoietin, vascular endothelial growth factor), and have been shown to deliver these therapeutic proteins either locally or systemically for months in vivo. Bone morphogenetic proteins belonging to the transforming growth factor-beta superfamily are osteoinductive molecules that drive the differentiation pathway of mesenchymal cells toward the chondroblastic or osteoblastic lineage, and stimulate bone formation in vivo. To determine whether skeletal muscle cells endogenously expressing bone morphogenetic proteins might serve as a vehicle for systemic bone morphogenetic protein delivery in vivo, proliferating skeletal myoblasts (C2C12) were transduced with a replication defective retrovirus containing the gene for recombinant human bone morphogenetic protein-6 (C2BMP-6). The C2BMP-6 cells constitutively expressed recombinant human bone morphogenetic protein-6 and synthesized bioactive recombinant human bone morphogenetic protein-6, based on increased alkaline phosphatase activity in coincubated mesenchymal cells. C2BMP-6 cells did not secrete soluble, bioactive recombinant human bone morphogenetic protein-6, but retained the bioactivity in the cell layer. Therefore, genetically-engineered skeletal muscle cells might serve as a platform for long-term delivery of osteoinductive bone morphogenetic proteins locally.

Review↗