Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Workload”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Flexible and Effective Object Tiering for Heterogeneous Memory Systems

Computing platforms that package multiple types of memory, each with their own performance characteristics, are quickly becoming mainstream. To operate efficiently, heterogeneous memory architectures require new data management solutions that are able to match the needs of each application with an appropriate type of memory. As the primary generators of memory usage, applications create a great deal of information that can be useful for guiding memory management, but the community still lacks tools to collect, organize, and leverage this information effectively. To address this gap, this work introduces a novel software framework that collects and analyzes object-level information to guide memory tiering. The framework includes tools to monitor the capacity and usage of individual data objects, routines that aggregate and convert this information into tier recommendations for the host platform, and mechanisms to enforce these recommendations according to user-selected policies. Moreover, the developed tools and techniques are fully automatic, work on standard Linux systems, and do not require modification or recompilation of existing software. Using this framework, this study evaluates and compares the impact of a variety of design choices for memory tiering, including different policies for prioritizing objects for the fast memory tier as well as the frequency and timing of migration events. In conclusion, the results, collected on a modern Intel platform with conventional DDR4 SDRAM as well as Intel Optane NVRAM, show that guiding data tiering with object-level information can enable significant performance and efficiency benefits compared with standard hardware- and software-directed data-tiering strategies for a diverse set of memory-intensive workloads.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

What to Support When You’re Compressing

Over the last nearly 20 years, lossy compression has become an essential aspect of HPC applications’ data pipelines, allowing them to overcome limitations in storage capacity and bandwidth and, in some cases, increase computational throughput and capacity. However, with the adoption of lossy compression comes the requirement to assess and control the impact lossy compression has on scientific outcomes. In this work, we take a major step forward in describing the state of practice and by characterizing workloads. We examine applications’ needs and compressors’ capabilities across 9 different supercomputing application domains. We present 24 takeaways that provide best practices for applications, operational impacts for facilities achieving compressed data, and gaps in application needs not addressed by production compressors that point towards opportunities for future compression research.

Error-Bounded Lossy Compression↗
DOE OSTI2025

RingX: Scalable Parallel Attention for Long-Context Learning on HPC

The attention mechanism has become foundational for remarkable AI breakthroughs since the introduction of the Transformer, driving the demand for increasingly longer context to power frontier models such as large-scale reasoning language models and high-resolution image/video generators. However, its quadratic computational and memory complexities present substantial challenges. Current state-of-the-art parallel attention methods, such as ring attention, are widely adopted for long-context training but utilize a point-to-point communication strategy that fails to fully exploit the capabilities of modern HPC network architectures. In this work, we propose ringX, a scalable family of parallel attention methods optimized explicitly for HPC systems. By enhancing workload partitioning, refining communication patterns, and improving load balancing, ringX achieves up to 3.4 × speedup compared to conventional ring attention on the Frontier supercomputer. Optimized for both bi-directional and causal attention mechanisms, ringX demonstrates its effectiveness through training benchmarks of a Vision Transformer (ViT) on a climate dataset and a Generative Pre-Trained Transformer (GPT) model, Llama3 8B. Our method attains an end-to-end training speedup of approximately 1.5 × in both scenarios. To our knowledge, the achieved 38% model FLOPs utilization (MFU) for training Llama3 8B with a 1M-token sequence length on 4,096 GPUs represents one of the highest training efficiencies reported for long-context learning on HPC systems. Our code implementation is available at https://github.com/jqyin/ringX-attention.

Yin, Junqi [ORNL] (ORCID:0000000338435520)↗
DOE OSTI2025

A Digital Twin of Scalable Quantum Clouds

Quantum computing has emerged as a transformative technology capable of solving complex problems beyond the limit of classical systems. The rapid development of quantum processors has led to the proliferation of cloud-based quantum computing services offered by platforms such as IBM, Google, and Amazon. These platforms introduce unique challenges in resource allocation, job scheduling, and multi-device orchestration as quantum workloads become increasingly complex. In this work, we present a digital twin of quantum cloud infrastructures: a framework designed to model and simulate the behavior of real quantum cloud systems. Developed in Python using the SimPy discrete-event simulation library, the framework replicates key aspects of quantum cloud environments, including detailed quantum device modeling, job lifecycle management, and job fidelity. It incorporates noise-aware fidelity estimation, making it the first of its kind to simulate superconducting gate-based quantum cloud systems at an administrative level with job fidelity. We present use cases as proof of concept, demonstrating that our quantum cloud simulation framework can act as a digital twin of a quantum cloud and support the modeling and implementation of practical systems.

Luo, Waylon [Kent State University]↗
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

In-Transit Data Transport Strategies for Coupled AI-Simulation Workflow Patterns

Coupled AI-Simulation workflows are becoming the major workloads for HPC facilities, and their increasing complexity necessitates new tools for performance analysis and prototyping of new in-situ workflows. We present SimAI-Bench, a tool designed to both prototype and evaluate these coupled workflows. In this paper, we use SimAI-Bench to benchmark the data transport performance of two common patterns on the Aurora supercomputer: a one-to-one workflow with co-located simulation and AI training instances, and a many-to-one workflow where a single AI model is trained from an ensemble of simulations. For the one-to-one pattern, our analysis shows that node-local and DragonHPC data staging strategies provide excellent performance compared Redis and Lustre file system. For the many-to-one pattern, we find that data transport becomes a dominant bottleneck as the ensemble size grows. Our evaluation reveals that file system is the optimal solution among the tested strategies for the many-to-one pattern.

Tummalapalli, Harikrishna [Argonne National Labora↗
DOE OSTI2026

T-FSM: A Scalable Distributed Task-Based System for Frequent Subgraph Pattern Mining from a Big Graph

Finding frequent subgraph patterns in a big graph is an important problem with many applications such as classifying chemical compounds and building indexes to speed up graph queries. Since this problem is NP-hard, some recent parallel and distributed systems have been developed to accelerate the mining. However, they often have a huge memory cost, very long running time, suboptimal load balancing, poor scale-out capability, and possibly inaccurate results. In this article, we propose an efficient system called T-FSM for parallel mining of frequent subgraph patterns in a big graph. T-FSM supports a new anti-monotonic frequentness measure called Fraction-Score, which is more accurate than the widely used MNI measure. The execution engine of T-FSM supports both intra-machine parallelism and inter-machine parallelism. For intra-machine parallelism, T-FSM adopts a novel task-based execution model to ensure high multithreading concurrency, bounded memory consumption, and effective load balancing. For inter-machine parallelism, T-FSM ensures good scale-out performance with a lightweight pattern rebalancing approach that reduces workload skewness of pattern evaluations among machines. To avoid recomputing the contexts for migrated patterns, we design a novel context cache table to support concurrent and asynchronous requesting and caching of remote context data, which can timely evict and garbage collect used pattern contexts that are no longer needed to keep memory consumption bounded. Extensive experiments show that T-FSM is orders of magnitude faster than existing state-of-the-art parallel systems (more than 10×, 51×, 131×, 55× speedup over ScaleMine, DistGraph, Pangolin and Peregrine, respectively) and distributed systems (more than 42× and 88× over ScaleMine and DistGraph, respectively) for frequent subgraph pattern mining, and it scales out satisfactorily to 512 CPU cores on the Polaris supercomputer at Argonne National Laboratory.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Wattchmen: Watching the Wattchers – High Fidelity, Flexible GPU Energy Modeling

Modern GPU-rich HPC systems are increasingly becoming energy-constrained. Thus, understanding an application’s energy consumption becomes essential. Unfortunately, current GPU energy attribution techniques are either inaccurate, inflexible, or outdated. Therefore, we propose Wattchmen, a flexible methodology for measuring, attributing, and predicting GPU energy consumption. We construct a per-instruction energy model using a diverse set of microbenchmarks to systematically quantify the energy consumption of GPU instructions, enabling finer-grain prediction and energy consumption breakdowns for applications. Compared with the state-of-the-art systems like AccelWattch (32%) and Guser (25%), across 16 popular GPGPU, graph analytics, HPC, and ML workloads, Wattchmen reduces the mean absolute percent error (MAPE) to 14% on V100 GPUs. Furthermore, we show that Wattchmen provides similar MAPEs for water-cooled V100s (15%) and extends to later architectures, including air-cooled A100 (11%) and H100 (12%) GPUs. Finally, to further demonstrate Wattchmen ’s value, we apply it to applications such as Backprop and QMCPACK, where Wattchmen ’s insights enable energy reductions of up to 35%.

Tran, Brandon [University of Wisconsin, Madison] (↗
DOE OSTI2025

High Velocity AI Cache (HVAC)

HVAC is a high-performance caching system designed for large-scale AI training workloads.

Khan, Awais [Oak Ridge National Laboratory (ORNL),↗
DOE OSTI2025

Adaptive Computing (AC) [SWR-24-106]

The Adaptive Computing (AC) software stack supports goal-based computing, for which a simulation workload is created on the fly adapting to the results of calculations. Application-specific code defines an objective, which may be to solve an optimization problem or to train a surrogate model with minimal uncertainty. Then, the AC driver decides where in the design parameter space to run simulations to best achieve that objective. This process is iterative and online; as new data is returned from simulations, the AC driver chooses new simulations to run. The AC driver can strategically run simulations on distributed hardware resources (including high performance computing machines, cloud resources, and edge devices) to maximize throughput and obey resource constraints.

Griffin, Kevin [National Renewable Energy Laborato↗
DOE OSTI2025

ESnet-JLab FPGA Accelerated Transport (control plane) [EJFAT (udplbd2)] v2.0

The ESnet-JLab FPGA Accelerated Transport system is a solution for streaming high-speed scientific measurement data from Data Acquisition Systems (DAQs) to high-performance computing facilties. It is generally compatible with many science workflows, and makes no assumptions about the specifics of any particular experiment. This program (udplbd version 2) implements the control plane for the system. It is responsible for programming network forwarding rules into the data plane (implemented by the hardware designed named udplb, described separately). It also implements the control loop necessary to match up offered workload with available capacity on high-performance compute nodes.

Howard, Derek [Lawrence Berkeley National Laborato↗
DOE OSTI2025

BULKI-Store v0.3.2

BULKI-Store is a distributed object storage system optimized for high-performance computing environments. Built with a Rust core and Python bindings, it efficiently manages scientific and machine learning datasets across HPC clusters. The system employs a client-server architecture with MPI integration, enabling seamless scaling on supercomputers like Perlmutter. BULKI-Store's object-oriented approach provides intuitive data organization with rich metadata support, contrasting with traditional file-based solutions. Key optimizations include selective checkpoint loading, unified checkpoint files, and object chunking for large data transfers. For machine learning workloads, BULKI-Store offers advantages through fine-grained access patterns, dynamic data sharing between training instances, and reduced memory pressure. Memory management features include strategic Python GC calls, minimized data copies, and batch processing capabilities. The system leverages Rayon's thread pool for asynchronous data prefetching and supports multiple CPU architectures (ARM64, x86, AMD, RISC-V). By combining performance optimizations with developer-friendly APIs, BULKI-Store addresses the complex data management challenges of modern HPC applications while maintaining compatibility across heterogeneous computing environments.

Zhang, Wei [Lawrence Berkeley National Laboratory ↗
DOE OSTI2025

Dori Tools v2.1.0

Dori Tools is a set of useful programs that support running a mid-range HPC cluster for BioInformatics workloads.

Rath, Georg [Lawrence Berkeley National Laboratory↗
DOE OSTI2026

NCCS High Performance GMRES Mixed Precision

HPG-MxP is a software package that performs a fixed number of multigrid preconditioned (using a Gauss-Seidel smoother) Generalized minimal residual (PGMRES) iterations in order to solve a possibly nonsymmetric large sparse linear system of equations. It is designed to be a benchmark to measure a computer's performance for sparse linear algebra workloads typical in scientific computing while allowing the use of mixed precision methods. The solution is required to have convergence characteristics and accuracy similar to double precision GMRES. It is based on the High Performance Conjugate Gradient Benchmark (HPCG) which restricts all implementations to use only the IEEE double precision format (FP64). The original implementation (https://github.com/hpg-mxp/hpg-mxp) was written by Ichitaro Yamazaki, Jennifer Loe, Christian Glusa, Sivasankaran Rajamanickam, Piotr Luszczek, and Jack Dongarra. Please refer to that repository for documentation on the original implementation. This version is maintained by the National Center for Computational Sciences at Oak Ridge National Laboratory. It is highly scalable and optimized for Oak Ridge Leadership Computing Facility (OLCF) systems, particularly Frontier.

Kashi, Aditya [Oak Ridge National Laboratory (ORNL↗
DOE OSTI2026

arco (Assembled Resource-Constrained Optimization) [SWR-26-030]

Arco (Assembled Resource-Constrained Optimization) is a memory-smart optimization DSL and solver for LP and MIP problems on constrained hardware. The software is an optimization framework built around a KDL-based domain-specific language and a CLI compiler/solver. You write optimization models in .kdl files, and the arco CLI compiles, validates, inspects, and solves them. Language bindings (Python today, more planned) provide programmatic access to the same engine. Built for harder optimization problems on constrained resources, Arco is intentional about every allocation, careful with stack and heap behavior, and relentless about minimizing memory usage so more systems can run real workloads. Arco is built primarily for internal use within our organization. You are welcome to try it, but we make no guarantees about API stability or robustness at this stage

Sanchez Perez, Pedro Andres [National Laboratory o↗
DOE OSTI2026

Supervisory Control and Data Acquisition for Electrochemical Separation Experimentation

The Python-based program is a laboratory automation tool designed to control and monitor electrochemical systems. The tool was developed for capacitive deionization (CDI) experiments, but it can be used for any system that requires controlled voltage or current segments and multi-parameter monitoring. The program integrates hardware components to run user-defined experimental parameters, providing operational control of a programmable power supply, peristaltic pump, and data acquisition devices. Currently, the program is structured with a workflow that includes an initialization (or pre-run) phase, a main loop, and a post-experiment stabilization (or post-run) phase. The initialization phase prepares and stabilizes the cell, ensuring that the electrodes and solution reach a baseline state before the experiment begins. The main loop consists of multiple voltage segments that repeat, controlling the experiment while recording key parameters such as time, voltage, current, pH, and conductivity. Finally, the post-experiment stabilization phase allows the system to stabilize after the experiment, returning the cell and solution to equilibrium conditions before ending the sequence. The program is designed with four variations, each tailored to different experimental needs. All variations include both the initialization and post-experiment stabilization stages, which run for a set amount of time, voltage, current, and flow rate before and after the main experiment block. The main loop runs for a set number of cycles, as defined by the user input, and each cycle is composed of 2 or 4 segments. The 4 program variations are described as follows: Program 1: The main program includes 2 segments. Each segment is defined to have a set duration, flow rate, voltage, and current. This program measures conductivity, flow rate, voltage, and current. Program 2: The main program expands Program 1 to include 4 segments. Each segment has a specified duration, flow rate, voltage, and current. Like Program 1, it measures conductivity, flow rate, voltage, and current. Program 3: The main program consists of 2 segments, each defined by time, flow rate, voltage, and current. In addition to conductivity, flow rate, voltage, and current, Program 3 collects pH and temperature data through a 4-channel data acquisition device. Program 4: This program independently controls two channels of a multi-channel power supply simultaneously. While conductivity can only be measured for one cell at a time, the dual-channel control makes it possible to operate two cells simultaneously under different voltage/current conditions. The main program includes 2 segments.For each program, all measurements are automatically logged and integrated into a single Excel output file. Data are displayed in numerical format and plotted, both in real time, to track system performance. A key feature of the program is its ability to synchronize all outputs so that every measurement shares a single timestamp, ensuring accurate alignment of voltage, current, pH, conductivity, and pH data.By combining hardware control, real-time monitoring, and unified data collection, this program significantly reduces manual workload and minimizes errors, making it a reliable platform for researchers, engineers, and laboratory technicians conducting CDI experiments, among other electrochemical tests.

Valentino, Lauren [Argonne National Laboratory (AN↗
DOE OSTI2026

Adaptive Computing (AC) (Open Source) [SWR-24-106]

The Adaptive Computing (AC) software stack supports goal-based computing, for which a simulation workload is created on the fly, adapting to the results of calculations. Application-specific code defines an objective, which may be to solve an optimization problem or to train a surrogate model with minimal uncertainty. Then, the AC driver decides where in the design parameter space to run simulations to best achieve that objective. This process is iterative and online; as new data is returned from simulations, the AC driver chooses new simulations to run. The AC driver can strategically run simulations on distributed hardware resources (including high performance computing machines, cloud resources, and edge devices) to maximize throughput and obey resource constraints.

Griffin, Kevin [National Laboratory of the Rockies↗
DOE OSTI2026

torch-einshard v1.0

torch-einshard is a Python library for describing local and distributed PyTorch tensor computations with compact, einsum-like notation. Its expressions name logical axes, specify how they are sharded across a PyTorch DeviceMesh, and represent partial reductions. The library automatically performs contractions, permutations, reshaping, splitting, gathering, reduction, reduce-scatter, and repartitioning while preserving autograd. Additional features include sharding-aware FFTs, tensor rolls, halo exchange, sliding windows, 1D–3D convolutions, uneven-shard handling, parameter initialization and gradient management, and cost-based execution planning. It is designed for scientific machine learning and large-model workloads, including tensor-, sequence-, and spatial-parallel MLPs, attention, convolutions, and spectral operations. Compared with manually combining torch.einsum and distributed collectives, torch-einshard expresses both the mathematical operation and data placement in one readable formula. This reduces boilerplate and synchronization errors, keeps forward and backward communication consistent, and allows the library to select optimized collective strategies without changing model code.

Morozov, Dmitriy [Lawrence Berkeley National Labor↗