Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “latency”

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 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training

Deep Learning Recommendation Models (DLRMs) are critical applications in various domains and have evolved as one of the single largest machine learning applications. Trillions of DLRM parameters exceed the on-chip memory capacity of GPUs. Large-scale multi-node systems are required for distributed DLRM inference and training, which suffer from the all-to-all communication bottleneck, mainly limiting the scalability of ever-growing DLRMs. In recent years, SmartNICs have evolved with coupled computation and communication capabilities providing opportunities for a powerful heterogeneous device in the system. However, there isn't such a distributed system that fully leverages the abundant smartNIC resources that resolve the scalability issue of DLRMs. In this work, we proposed a software-hardware co-design of a heterogeneous smartNIC system that resolves the communication bottleneck of distributed DLRMs, mitigates the memory bandwidth pressure, and improves computation efficiency. We provide a set of smartNIC designs of cache systems (including local cache and remote cache) and smartNIC computation kernels which reduce data movement, relieve memory lookup intensity, and improve the GPU's computation efficiency. In addition, we propose a graph algorithm that improves the data locality of queries within batches which optimizes the overall system performance with higher data reuse. Our evaluation shows that our system achieves 2.1x latency speedup for inference and 1.6x throughput speedup for training.

Guo, Anqi↗
DOE OSTI2023

Unified Communication Optimization Strategies for Sparse Triangular Solver on CPU and GPU Clusters

This paper presents a unified communication optimization framework for sparse triangular solve (SpTRSV) algorithms on CPU and GPU clusters. The framework builds upon a 3D communication-avoiding (CA) layout of Px × Py × Pz processes that divides a sparse matrix into Pz submatrices, each handled by a Px × Py 2D grid with block-cyclic distribution. We propose three communication optimization strategies: First, a new 3D SpTRSV algorithm is developed, which trades the inter-grid communication and synchronization with replicated computation. This design requires only one inter-grid synchronization, and the inter-grid communication is efficiently implemented with sparse allreduce operations. Second, broadcast and reduction communication trees are used to reduce message latency of the intra-grid 2D communication on CPU clusters. Finally, we leverage GPU-initiated one-sided communication to implement the communication trees on GPU clusters. With these nested inter- and intra-grid communication optimization strategies, the proposed 3D SpTRSV algorithm can attain up to 3.45x speedups compared to the baseline 3D SpTRSV algorithm using up to 2048 Cori Haswell CPU cores. In addition, the proposed GPU 3D SpTRSV algorithm can achieve up to 6.5x speedups compared to the proposed CPU 3D SpTRSV algorithm with Pz up to 64. Finally it is remarkable that the proposed GPU 3D SpTRSV can scale to 256 GPUs using the Perlmutter system while the existing 2D SpTRSV algorithm can only scale up to 4 GPUs.

Sao, Piyush↗
DOE OSTI2023

QuComm: Optimizing Collective Communication for Distributed Quantum Computing

Distributed quantum computing (DQC) is a scalable way to build a large-scale quantum computing system while the error-prone nonlocal communication between DQC nodes may heavily degrade the fidelity of the distributed quantum program and thus demands specific compiler optimizations. Previous compilers on DQC communication optimization either assumes unlimited communication resource or a few communication qubits due to the hardware limitation. The former compilers may not be efficient when interfacing with communication-resource-constrained DQC hardware while the latter compilers lose the opportunities of optimizing collective communication and routing concurrent communication as they unnecessarily couple limited communication qubits with the implementation of expensive inter-node operations. In this paper, we invent the communication buffer, a communication facility consisting of idle qubits in each compute node, to decouple the execution of inter-node quantum operations from communication qubits: communication qubits are devoted to generating inter-node entanglement while internode operations are conducted in the communication buffer. The communication buffer provides an intermediate layer for inter-node communication and paves the way for collective communication optimization. We then propose QuComm, a buffer-based compiler framework that first performs smart buffer allocation according to communication characteristics of the distributed quantum program and then optimizes and collectively routes inter-node quantum operations. Experimental results on a hierarchical DQC system show that the proposed QuComm can reduce the most expensive inter-node communication request and the latency of various distributed quantum programs by 50.4% and 47.6% on average, respectively.

Wu, Anbang↗
DOE OSTI2023

DDStore: Distributed Data Store for Scalable Training of Graph Neural Networks on Large Atomistic Modeling Datasets

Graph neural networks (GNNs) are a class of Deep Learning models used in designing atomistic materials for effective screening of large chemical spaces. To ensure robust prediction, GNN models must be trained on large volumes of atomistic data on leadership class supercomputers. Even with the advent of modern architectures that consist of multiple storage layers that include node-local NVMe devices in addition to device memory for caching large datasets, extreme-scale model training faces I/O challenges at scale.We present DDStore, an in-memory distributed data store designed for GNN training on large-scale graph data. DDStore provides a hierarchical, distributed, data caching technique that combines data chunking, replication, low-latency random access, and high throughput communication. DDStore achieves near-linear scaling for training a GNN model using up to 1000 GPUs on the Summit and Perlmutter supercomputers, and reaches up to a 6.15x reduction in GNN training time compared to state-of-the-art methodologies.

Choi, Jong Youl↗
DOE OSTI2024

Cookie-Jar: An Adaptive Re-configurable Framework for Wireless Network Infrastructures

5G advancements like Massive Multiple Input Multiple Output (MIMO) bring high capacity and low latency, but also intensify interference challenges. Static and dynamic coordination techniques address this, often at the cost of increased power draw. We introduce Cookie-Jar (CJ), an interference coordination (IC) framework using reinforcement learning for multi-goal optimization. By dynamically adjusting network, power, and topology parameters based on real-time conditions, CJ improves Signal to Noise and Interference Ratio (SINR) while minimizing power consumption. Simulated 5G experiments showcase CJ's potential, achieving a 15% SINR improvement with near-identical power draw compared to existing methods.

Network↗
DOE OSTI2024

CommBench: Micro-Benchmarking Hierarchical Networks with Multi-GPU, Multi-NIC Nodes

Modern high-performance computing systems have multiple GPUs and network interface cards (NICs) per node. The resulting network architectures have multilevel hierarchies of subnetworks with different interconnect and software technologies. These systems offer multiple vendor-provided communication capabilities and library implementations (IPC, MPI, NCCL, RCCL, OneCCL) with APIs providing varying levels of performance across the different levels. Understanding this performance is currently difficult because of the wide range of architectures and programming models (CUDA, HIP, OneAPI). We present CommBench, a library with cross-system portability and a high-level API that enables developers to easily build microbenchmarks relevant to their use cases and gain insight into the performance (bandwidth & latency) of multiple implementation libraries on different networks. We demonstrate CommBench with three sets of microbenchmarks that profile the performance of six systems. Our experimental results reveal the effect of multiple NICs on optimizing the bandwidth across nodes and also present the performance characteristics of four available communication libraries within and across nodes of NVIDIA, AMD, and Intel GPU networks.

Hidayetoglu, Mert↗
DOE OSTI2024

End-to-end codesign of Hessian-aware quantized neural networks for FPGAs

Here, we develop an end-to-end workflow for the training and implementation of co-designed neural networks (NNs) for efficient field-programmable gate array (FPGA) hardware. Our approach leverages Hessian-aware quantization of NNs, the Quantized Open Neural Network Exchange intermediate representation, and the hls4ml tool flow for transpiling NNs into FPGA firmware. This makes efficient NN implementations in hardware accessible to nonexperts in a single open sourced workflow that can be deployed for real-time machine-learning applications in a wide range of scientific and industrial settings. We demonstrate the workflow in a particle physics application involving trigger decisions that must operate at the 40-MHz collision rate of the CERN Large Hadron Collider (LHC). Given the high collision rate, all data processing must be implemented on FPGA hardware within the strict area and latency requirements. Based on these constraints, we implement an optimized mixed-precision NN classifier for high-momentum particle jets in simulated LHC proton-proton collisions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

SPARTA: High-Level Synthesis of Parallel Multi-Threaded Accelerators

This article presents a methodology for the Synthesis of PARallel multi-Threaded Accelerators (SPARTA) from OpenMP annotated C/C++ specifications. SPARTA extends an open-source HLS tool, enabling the generation of accelerators that provide latency tolerance for irregular memory accesses through multithreading, support fine-grained memory-level parallelism through a hot-potato deflection-based network-on-chip (NoC), support synchronization constructs, and can instantiate memory-side caches. Our approach is based on a custom runtime OpenMP library, providing flexibility and extensibility. Experimental results show high scalability when synthesizing irregular graph kernels. The accelerators generated with our approach are, on average, 2.29x faster than state-of-the-art HLS methodologies.

Design automation↗
DOE OSTI2025

From Edge to HPC: Investigating Cross-Facility Data Streaming Architectures

In this paper, we investigate three cross-facility data streaming architectures, Direct Streaming (DTS), Proxied Streaming (PRS), and Managed Service Streaming (MSS). We examine their architectural variations in data flow paths and deployment feasibility, and detail their implementation using the Data Streaming to HPC (DS2HPC) architectural framework and the SciStream memory-to-memory streaming toolkit on the production-grade Advanced Computing Ecosystem (ACE) infrastructure at Oak Ridge Leadership Computing Facility (OLCF). We present a workflow-specific evaluation of these architectures using three synthetic workloads derived from the streaming characteristics of scientific workflows. Through simulated experiments, we measure streaming throughput, round-trip time, and overhead under work sharing, work sharing with feedback, and broadcast and gather messaging patterns commonly found in AI-HPC communication motifs. Our study shows that DTS offers a minimal-hop path, resulting in higher throughput and lower latency, whereas MSS provides greater deployment feasibility and scalability across multiple users but incurs significant overhead. PRS lies in between, offering a scalable architecture whose performance matches DTS in most cases.

George, Anjus [ORNL] (ORCID:0000000179737061)↗
DOE OSTI2025

SlimIO: Lightweight I/O Path Design for Write Isolation in FDP-backed In-Memory Databases

In-Memory Databases (IMDBs) are widely used with HPC applications to manage transient data, often using snapshot-based persistence for backups. Redis, a representative IMDB, employs both snapshot and Write-Ahead Log (WAL) mechanisms, storing data on persistent devices via the traditional kernel I/O path. This method incurs syscall overhead, I/O contention between processes, and SSD garbage collection (GC) delays. To address these issues, we propose SlimIO, which adopts I/O passthru to minimize syscall overhead and inter-process I/O interference. Additionally, it leverages Flexible Data Placement (FDP) SSDs as backup storage to avoid performance degradation from SSD GC. Experimental results show that SlimIO reduces snapshot time by up to 25%, increases query throughput by up to 30% during non-snapshot periods, and lowers 99.9%-ile latency by up to 50%. Furthermore, it achieves a write amplification factor (WAF) of 1.00, indicating no redundant internal writes, thus extending SSD lifespan.

Lee, Sangyun [Sogang University]↗
DOE OSTI2025

Architecture-Aware Models of AI Engines for High-Performance Matrix Matrix Multiplication

The AI Engine (AIE) architecture, available in systems from mobile SoCs to server-class FPGAs, aims to efficiently execute AI/ML tasks through a two-dimensional array of compute tiles. Previous work on AIEs has explored different approaches to mapping computation across spatial arrays, but the compute kernel running on each tile has not been the focus. Additionally, the AIE-ML architecture introduces memory tiles and omits programmable logic, requiring new approaches to staging and moving data throughout the array. In this work we update analytical models developed for CPUs to produce the design of high performance kernels while introducing new model considerations such as memory structure, throughput, and latency as required by the AIE hardware. We evaluate our models by developing AIE-ML kernels for matrix multiplication in low-precision data types showing performance up to 95% of compute peak for the kernel when data resides in local memory and above 90% of compute peak when data resides in main memory.

Binder, Elliott D. [Carnegie Mellon University, Pi↗
DOE OSTI2026

CORE-BFS: Communication-Optimized REctangular-partitioned BFS Achieving 160.845 TeraTEPS on Frontier Supercomputer

Distributed Breadth-First Search (BFS) is fundamental to many large-scale graph applications, but its performance on parallel systems is often limited by high communication overhead. This paper presents CORE-BFS, an extremely scalable GPU-based BFS implementation that introduces a unique rectangular 2D partitioning-based design for Frontier supercomputer. To further improve performance, we propose four key optimizations: (1) Rectangular 2D-partition specific data formats that use two compressed row and one compressed column status array bitmaps combined with a Double Compressed Sparse Row (DCSR) format per partition, reducing memory footprint and inter-rank traffic; (2) Adaptive frontier & communication strategy that unifies top-down and bottom-up traversal on the rectangular layout, uses lazy synchronization in top-down levels, and switches variants based on frontier size to minimize communication overhead; (3) Frontier-split degree-aware update that maps frontier vertices to thread-centric, wavefront-centric, and block-centric kernels based on their degree to improve GPU utilization and memory coalescing; (4) Row-reduction pipeline that overlaps bottom-up adjacency list processing with row-wise bitmap reduction to hide inter-rank latency. Together, these techniques increase parallelism while reducing memory and communication overhead. On the Graph500 benchmark, CORE - BFS scales up to 9,248 Frontier nodes with scale-42 graphs and reaches 160.845 TTEPS, delivering a 5.42 × speedup over our previous Frontier implementation.

Yang, Haoshen [Rutgers University]↗
DOE OSTI2021

An Intelligent Garbage Sorting System Based on Edge Computing and Visual Understanding of Social Internet of Vehicles

In order to enable Social Internet of Vehicles devices to achieve the purpose of intelligent and autonomous garbage classification in a public environment, while avoiding network congestion caused by a large amount of data accessing the cloud at the same time, it is therefore considered to combine mobile edge computing with Social Internet of Vehicles to give full play to mobile edge computing features of high bandwidth and low latency. At the same time, based on cutting-edge technologies such as deep learning, knowledge graph, and 5G transmission, the paper builds an intelligent garbage sorting system based on edge computing and visual understanding of Social Internet of Vehicles. First of all, for the massive multisource heterogeneous Social Internet of Vehicles big data in the public environment, different item modal data adopts different processing methods, aiming to obtain a visual understanding model. Secondly, using the 5G network, the model is deployed on the edge device and the cloud for cloud-side collaborative management, aiming to avoid the waste of edge node resources, while ensuring the data privacy of the edge node. Finally, the Social Internet of Vehicles devices is used to make intelligent decision-making on the big data of the items. First, the items are judged as garbage, and then the category is judged, and finally the task of grabbing and sorting is realized. The experimental results show that the system proposed in this paper can efficiently process the big data of Social Internet of Vehicles and make valuable intelligent decisions. At the same time, it also has a certain role in promoting the promotion of Social Internet of Vehicles devices.

Shen, Xuehao↗
DOE OSTI2022

Effect of Low Frequency Repetitive Transcranial Magnetic Stimulation (rTMS) Combined with Hyperbaric Oxygen (HBO) on Awakening of Coma Patients with Traumatic Brain Injury

Coma caused by craniocerebral injury is a common condition of neurosurgical acute injury. There is no specific method to promote awakening in a clinic. Early comprehensive treatment may be helpful to patients. The common methods are hyperbaric oxygen (HBO) and low-frequency repetitive transcranial magnetic stimulation (rTMS). However, the application effect and mechanism of rTMS combined with HBO on coma patients with traumatic brain injury need to be further studied. The brain stem auditory evoked potential (BAEP) is examined by the Kennedy coma recovery scale (CRS-R), the recovery of brain function and the state of consciousness are evaluated, and the therapeutic effect is evaluated by the Glasgow Coma Scale (GCS). Cerebrospinal fluid NE level, MCA blood flow velocity, and left brainstem and right brainstem auditory evoked potential are used to evaluate brain rehabilitation. RTMS combined with HBO could shorten the wake-up time, improve the wake-up rate, improve the GCS score and CRS-R score, shorten the brain wave latency time of the left and right brainstem, increase the NE level of cerebrospinal fluid, and decrease the blood flow velocity of MCA. RTMS combines with HBO can improve the nerve excitability of brain cells, reduce the disturbance of consciousness, promote the functional recovery of brain injury, and has a certain role in promoting the awakening of patients with traumatic brain injury coma.

Liu, Mei↗
DOE OSTI2021

Analytical Tool to evaluate Heterogeneous Neuromorphic Architectures (ATHENA)

SAND2022-7062 O The Analytical Tool to Evaluate Heterogeneous Neuromorphic Architectures (ATHENA) quickly evaluates performance metrics like energy, area, and latency for an AI/ML network. It currently supports the evaluation of analog neural networks. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Hahn, Andrew↗
DOE OSTI2026

wa-hls4ml-paper

Code for plots, models, data generation and other utilities relating to the paper "wa-hls4ml: A Benchmark and Surrogate Models for hls4ml Resource and Latency Estimation" https://arxiv.org/abs/2511.05615 [FERMILAB-PUB-25-0359-CSAID]

Hawks, Ben [Fermi National Accelerator Laboratory ↗
DOE OSTI2024

Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications

Abstract Robust quantification of predictive uncertainty is a critical addition needed for machine learning applied to weather and climate problems to improve the understanding of what is driving prediction sensitivity. Ensembles of machine learning models provide predictive uncertainty estimates in a conceptually simple way but require multiple models for training and prediction, increasing computational cost and latency. Parametric deep learning can estimate uncertainty with one model by predicting the parameters of a probability distribution but does not account for epistemic uncertainty. Evidential deep learning, a technique that extends parametric deep learning to higher-order distributions, can account for both aleatoric and epistemic uncertainties with one model. This study compares the uncertainty derived from evidential neural networks to that obtained from ensembles. Through applications of the classification of winter precipitation type and regression of surface-layer fluxes, we show evidential deep learning models attaining predictive accuracy rivaling standard methods while robustly quantifying both sources of uncertainty. We evaluate the uncertainty in terms of how well the predictions are calibrated and how well the uncertainty correlates with prediction error. Analyses of uncertainty in the context of the inputs reveal sensitivities to underlying meteorological processes, facilitating interpretation of the models. The conceptual simplicity, interpretability, and computational efficiency of evidential neural networks make them highly extensible, offering a promising approach for reliable and practical uncertainty quantification in Earth system science modeling. To encourage broader adoption of evidential deep learning, we have developed a new Python package, Machine Integration and Learning for Earth Systems (MILES) group Generalized Uncertainty for Earth System Science (GUESS) (MILES-GUESS) ( https://github.com/ai2es/miles-guess ), that enables users to train and evaluate both evidential and ensemble deep learning. Significance Statement This study demonstrates a new technique, evidential deep learning, for robust and computationally efficient uncertainty quantification in modeling the Earth system. The method integrates probabilistic principles into deep neural networks, enabling the estimation of both aleatoric uncertainty from noisy data and epistemic uncertainty from model limitations using a single model. Our analyses reveal how decomposing these uncertainties provides valuable insights into reliability, accuracy, and model shortcomings. We show that the approach can rival standard methods in classification and regression tasks within atmospheric science while offering practical advantages such as computational efficiency. With further advances, evidential networks have the potential to enhance risk assessment and decision-making across meteorology by improving uncertainty quantification, a longstanding challenge. This work establishes a strong foundation and motivation for the broader adoption of evidential learning, where properly quantifying uncertainties is critical yet lacking.

Schreck, John S.↗
DOE OSTI2020

Near-Real-Time Forecast of Satellite-Based Soil Moisture Using Long Short-Term Memory with an Adaptive Data Integration Kernel

Nowcasts, or near-real-time (NRT) forecasts, of soil moisture based on the Soil Moisture Active and Passive (SMAP) mission could provide substantial value for a range of applications including hazards monitoring and agricultural planning. To provide such a NRT forecast with high fidelity, we enhanced a time series deep learning architecture, long short-term memory (LSTM), with a novel data integration (DI) kernel to assimilate the most recent SMAP observations as soon as they become available. The kernel is adaptive in that it can accommodate irregular observational schedules. Testing over the CONUS, this NRT forecast product showcases predictions with unprecedented accuracy when evaluated against subsequent SMAP retrievals. It showed smaller error than NRT forecasts reported in the literature, especially at longer forecast latency. The comparative advantage was due to LSTM’s structural improvements, as well as its ability to utilize more input variables and more training data. The DI-LSTM was compared to the original LSTM model that runs without data integration, referred to as the projection model here. We found that the DI procedure removed the autocorrelated effects of forcing errors and errors due to processes not represented in the inputs, for example, irrigation and floodplain/lake inundation, as well as mismatches due to unseen forcing conditions. The effects of this purely data-driven DI kernel are discussed for the first time in the geosciences. Furthermore, this work presents an upper-bound estimate for the random component of the SMAP retrieval error.

54 ENVIRONMENTAL SCIENCES↗