Search NASA⌕ Search

SEARCH · Search NASA

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 793 records · Page 44

Modeling Large-Scale Slim Fly Networks Using Parallel Discrete-Event Simulation

As supercomputers approach exascale performance, the increased number of processors translates to an increased demand on the underlying network interconnect. We present that the slim fly network topology, a new low-diameter, low-latency, and low-cost interconnection network, is gaining interest as one possible solution for next-generation supercomputing interconnect systems. In this article, we present a high-fidelity slim fly packet-level model leveraging the Rensselaer Optimistic Simulation System (ROSS) and Co-Design of Exascale Storage (CODES) frameworks. We validate the model with published work before scaling the network size up to an unprecedented 1 million compute nodes and confirming that the slim fly observes peak network throughput at extreme scale. In addition to synthetic workloads, we evaluate large-scale slim fly models with real communication workloads from applications in the Design Forward program with over 110,000 MPI processes. We show strong scaling of the slim fly model on an Intel cluster achieving a peak network packet transfer rate of 2.3 million packets per second and processing over 7 billion discrete events using 128 MPI tasks. Enabled by the strong performance capabilities of the model, we perform a detailed application trace and routing protocol performance study. Lastly, through analysis of metrics such as packet latency, hop count, and congestion, we find that the slim fly network is able to leverage simple minimal routing and achieve the same performance as more complex adaptive routing for tested DOE benchmark applications.

97 MATHEMATICS AND COMPUTING↗

LADR: low-cost application-level detector for reducing silent output corruptions

Applications running on future high performance computing (HPC) systems are more likely to experience transient faults due to technology scaling trends with respect to higher circuit density, smaller transistor size and near-threshold voltage (NTV) operations. A transient fault could corrupt application state without warning, possibly leading to incorrect application output. Such errors are called silent data corruptions (SDCs). In this paper, we present LADR, a low-cost application-level SDC detector for scientific applications. LADR protects scientific applications from SDCs by watching for data anomalies in their state variables (those of scientific interest). It employs compile-time data-flow analysis to minimize the number of monitored variables, thereby reducing runtime and memory overheads while maintaining a high level of fault coverage with low false positive rates. We evaluated LADR with 4 scientific workloads and results show that LADR achieved < 80% fault coverage with only ∼ 3% runtime overheads and ∼ 1% memory overheads. As compared to prior state-of-the-art anomaly-based detection methods, SDC achieved comparable or improved fault coverage, but reduced runtime overheads by 21% ∼ 75%, and memory overheads by 35% ∼ 55% for the evaluated workloads. We believe that such an approach with low memory and runtime overheads coupled with attractive detection precision makes LADR a viable approach for assuring the correct output from large-scale high performance simulations.

Chen, Chao↗

The Minos Computing Library: Efficient Parallel Programming for Extremely Heterogeneous Systems

Hardware specialization has become the silver bullet to achieve efficient high performance, from Systems-on-Chip systems, where hardware specialization can be ``extreme'', to large-scale HPC systems. As the complexity of the systems increases, so does the complexity of programming such architectures in a portable way. This work introduces the Minos Computing Library (MCL), as system software, programming model, and programming model runtime that facilitate programming extremely heterogeneous systems. MCL supports the execution of several multi-threaded applications within the same compute node, performs asynchronous execution of application tasks, efficiently balances computation across hardware resources, and provides performance portability. We show that code developed on a personal desktop automatically scales up to fully utilize powerful workstations with 8 GPUs and down to power-efficient embedded systems. MCL provides up to 17.5x speedup over OpenCL on NVIDIA DGX-1 systems and up to 1.88x speedup on single-GPU systems. In multi-application workloads, MCL dynamically resource allocation provides up to 2.43x performance improvement over manual, static allocation of computing resources.

Gioiosa, Roberto↗

DFSynthesizer: Dataflow-based Synthesis of Spiking Neural Networks to Neuromorphic Hardware

Spiking Neural Networks (SNNs) are an emerging computation model that uses event-driven activation and bio-inspired learning algorithms. SNN-based machine learning programs are typically executed on tile-based neuromorphic hardware platforms, where each tile consists of a computation unit called a crossbar, which maps neurons and synapses of the program. However, synthesizing such programs on an off-the-shelf neuromorphic hardware is challenging. This is because of the inherent resource and latency limitations of the hardware, which impact both model performance, e.g., accuracy, and hardware performance, e.g., throughput. We propose DFSynthesizer, an end-to-end framework for synthesizing SNN-based machine learning programs to neuromorphic hardware. The proposed framework works in four steps. First, it analyzes a machine learning program and generates SNN workload using representative data. Second, it partitions the SNN workload and generates clusters that fit on crossbars of the target neuromorphic hardware. Third, it exploits the rich semantics of the Synchronous Dataflow Graph (SDFG) to represent a clustered SNN program, allowing for performance analysis in terms of key hardware constraints such as number of crossbars, dimension of each crossbar, buffer space on tiles, and tile communication bandwidth. Finally, it uses a novel scheduling algorithm to execute clusters on crossbars of the hardware, guaranteeing hardware performance. We evaluate DFSynthesizer with 10 commonly used machine learning programs. Our results demonstrate that DFSynthesizer provides a much tighter performance guarantee compared to current mapping approaches.

Computer Science↗

Exploring the Use of Novel Spatial Accelerators in Scientific Applications

Driven by the need to find alternative accelerators which can viably replace GPUs in next-generation Supercomputing systems, this paper proposes a methodology to enable agile application/hardware co-design. The application-first methodology provides the ability to come up with design of accelerators while working with real-world workloads, available accelerators, and system software. The iterative design process targets a set of kernels in a workload for performance estimates that can prune the design space for later phases of detailed architectural evaluations. To this effect, in this paper, a novel data-parallel device model is introduced that simulates the latency of performance-sensitive operations in an accelerator including data transfers and kernel computation using multi-core CPUs. The use of off-the-shelf simulators, such as pre-RTL simulator Aladdin or multiple tools available for exploring the design of deep neural network accelerators (e.g., Timeloop) is demonstrated for evaluation of various accelerator designs using applications with realistic inputs. Examples of multiple device configurations that are instantiable in a system are explored to evaluate the performance benefit of deploying novel accelerators. The proposed device is integrated with a programming model and system software to potentially explore the impacts of high-level programming languages/compilers and low-level effects such as task scheduling on multiple accelerators. We analyze our methodology for a set of applications that represent high-performance computing (HPC) and graph analytics. The applications include a computational chemistry kernel realized using tensor contractions, triangle counting, GraphSAGE and Breadth-first Search. These applications include kernels such as dense matrix-dense matrix multiplication, sparse matrix-spare matrix multiplication, and sparse matrix-dense vector multiplication. Our results indicate potential performance benefits and insights for system design by including accelerators that realize these kernels along-side general purpose accelerators.

AI, codesign, Accelerated Computing, Modeling and ↗

Access Patterns and Performance Behaviors of Multi-layer Supercomputer I/O Subsystems under Production Load

Scientific computing workloads at HPC facilities have been shifting from traditional numerical simulations to AI/ML applications for training and inference while processing and producing ever-increasing amounts of scientific data. To address the growing need for increased storage capacity, lower access latency, and higher bandwidth, emerging technologies such as non-volatile memory are integrated into supercomputer I/O subsystems. With these emerging trends, we need a better understanding of the multilayer supercomputer I/O systems and ways to use these subsystems efficiently. In this work, we study the I/O access patterns and performance characteristics of two representative supercomputer I/O subsystems. Through an extensive analysis of year-long I/O logs on each system, we report new observations in I/O reads and writes, unbalanced use of storage system layers, and new trends in user behaviors at the HPC I/O middleware stack.

Bez, JL↗

A Case For Intra-rack Resource Disaggregation in HPC

The expected halt of traditional technology scaling is motivating increased heterogeneity in high-performance computing (HPC) systems with the emergence of numerous specialized accelerators. As heterogeneity increases, so does the risk of underutilizing expensive hardware resources if we preserve today’s rigid node configuration and reservation strategies. This has sparked interest in resource disaggregation to enable finer-grain allocation of hardware resources to applications. However, there is currently no data-driven study of what range of disaggregation is appropriate in HPC. To that end, we perform a detailed analysis of key metrics sampled in NERSC’s Cori, a production HPC system that executes a diverse open-science HPC workload. In addition, we profile a variety of deep-learning applications to represent an emerging workload. We show that for a rack (cabinet) configuration and applications similar to Cori, a central processing unit with intra-rack disaggregation has a 99.5% probability to find all resources it requires inside its rack. In addition, ideal intra-rack resource disaggregation in Cori could reduce memory and NIC resources by 5.36% to 69.01% and still satisfy the worst-case average rack utilization.

97 MATHEMATICS AND COMPUTING↗

FLASH: FPGA-Accelerated Smart Switches with GCN Case Study

Some communication switches, e.g., the Mellanox SHArP and those in the IBM BlueGene clusters, are augmented to process packets at the application level with fixed-function collectives. This approach, however, lacks flexibility, which limits their applicability in diverse and dynamic workloads. Recently, a new type of programmable packet processor, which uses high-level languages, e.g., P4, has emerged as possible candidates. P4-based switches, however, fall short in certain applications, including machine learning, where capabilities not currently supported by P4 are needed. These include more complex calculation, such as sparse computation and fused multiply-accumulate, data-intensive floating point operations, data reuse, and significant memory. The problem addressed here is that such a switch augmentation needs to support: a large amount of state, significant flexible compute capability, and ease of programming, all while maintaining full functionality, including ensuring high throughput, and demonstrating utility. In this work, we propose a programmable look-aside-type accelerator that can be embedded into, or attached to, existing communication switch pipelines and that is capable of processing packets at line-rate. The proposed in-switch accelerator is based on mixing an ISA (subset of RISC-V instructions) with dataflow graphs (found in CGRAs). To augment performance, vector instructions are also supported. To facilitate usability, we have developed a complete toolchain to compile user-provided C/C++ codes to appropriate back-end instructions for configuring the accelerator. While this approach is flexible enough to support various workloads, in this paper, we consider Graph Convolutional Networks (GCNs) as a case study. Experimental results show that this approach considerably improves the performance of distributed GCN applications.

Haghi, Pouya↗

Evaluating Emerging AI/ML Accelerators: IPU, RDU, and and NVIDIA/AMD GPUs

As the size and complexity of AI/ML workloads continue to increase, the demand for specialized hardware accelerators has grown rapidly. To better understand the landscape of commercial AI/ML accelerators, we evaluate and compare several popular platforms, including Graphcore IPU, Sambanova RDU, and GPUs, on a range of AI workloads. Our research aims to help researchers and developers understand the unique features of commercial AI/ML accelerators and provide reference design and performance numbers for research prototypes. By shedding light on the current landscape of commercial AI/ML accelerators, we hope to offer insights into the future development of specialized hardware accelerators for AI/ML.

Peng, Hongwu↗

Parallel Simulation of Quantum Networks with Distributed Quantum State Management

Quantum network simulators offer the opportunity to cost-efficiently investigate potential avenues for building networks that scale with the number of users, communication distance, and application demands by simulating alternative hardware designs and control protocols. Several quantum network simulators have been recently developed with these goals in mind. As the size of the simulated networks increases, however, sequential execution becomes time-consuming. Parallel execution presents a suitable method for scalable simulations of large-scale quantum networks, but the unique attributes of quantum information create unexpected challenges. In this work, we identify requirements for parallel simulation of quantum networks and develop the first parallel discrete-event quantum network simulator by modifying the existing serial simulator SeQUeNCe. Our contributions include the design and development of a quantum state manager (QSM) that maintains shared quantum information distributed across multiple processes. We also optimize our parallel code by minimizing the overhead of the QSM and decreasing the amount of synchronization needed among processes. Using these techniques, we observe a speedup of 2 to 25 times when simulating a 1,024-node linear network topology using 2 to 128 processes. We also observe an efficiency greater than 0.5 for up to 32 processes in a linear network topology of the same size and with the same workload. We repeat this evaluation with a randomized workload on a caveman network. We also introduce several methods for partitioning networks by mapping them to different parallel simulation processes. We have released the parallel SeQUeNCe simulator as an open source tool alongside the existing sequential version.

97 MATHEMATICS AND COMPUTING↗

DEDUPKV: A Space-Efficient and High-Performance Key-Value Store via Fine-Grained Deduplication

Log-Structured Merge Tree (LSM-tree) based key-value stores excel in write-intensive environments but suffer from data duplication, consuming up to 49% of storage space in LSM-tree-based key-value store deployments. Traditional solutions like compression and coarse-grained file system-level deduplication introduce overhead or have limited effectiveness. In this study, we propose DedupKV, a fine-grained deduplication framework tailored for LSM-tree, maximizing data reduction efficiency while minimizing write stalls and read overheads. DedupKV features three key innovations: (1) FLUSH-integrated inline deduplication, which removes duplicates during memory-to-storage writes; (2) WAL file-based offline deduplication, repurposing write-ahead logs to avoid double writes; and (3) elastic execution, dynamically balancing inline and offline deduplication based on memory pressure and workload intensity. Additionally, dynamic granularity management reduces deduplication metadata overhead. We implemented these four ideas in RocksDB for the first time and conducted experiments in a Linux environment. Our evaluation shows that WAL file-based offline deduplication and DedupKV outperform BlobDB by 33% and 23%, respectively, in write-heavy workloads, while reducing write amplification by 1.2 ×, 2 ×, and 1.6 × for real KV datasets.

Jamil, Safdar [Sogang University]↗

I/O in Machine Learning Applications on HPC Systems: A 360-degree Survey

Growing interest in Artificial Intelligence (AI) has resulted in a surge in demand for faster methods of Machine Learning (ML) model training and inference. This demand for speed has prompted the use of high performance computing (HPC) systems that excel in managing distributed workloads. Because data is the main fuel for AI applications, the performance of the storage and I/O subsystem of HPC systems is critical. In the past, HPC applications accessed large portions of data written by simulations or experiments or ingested data for visualizations or analysis tasks. ML workloads perform small reads spread across a large number of random files. This shift of I/O access patterns poses several challenges to modern parallel storage systems. In this paper, we survey I/O in ML applications on HPC systems, and target literature within a 6-year time window from 2019 to 2024. We define the scope of the survey, provide an overview of the common phases of ML, review available profilers and benchmarks, examine the I/O patterns encountered during offline data preparation, training, and inference, and explore I/O optimizations utilized in modern ML frameworks and proposed in recent literature. Lastly, we seek to expose research gaps that could spawn further R&D.

97 MATHEMATICS AND COMPUTING↗

Mojo: MLIR-based Performance-Portable HPC Science Kernels on GPUs for the Python Ecosystem

We explore the performance and portability of the novel Mojo language for scientific computing workloads on GPUs. As the first language based on the LLVM’s Multi-Level Intermediate Representation (MLIR) compiler infrastructure, Mojo aims to close performance and productivity gaps by combining Python’s interoperability and CUDA-like syntax for compile-time portable GPU programming. We target four scientific workloads: a seven-point stencil (memory-bound), BabelStream (memory-bound), miniBUDE (compute-bound), and Hartree–Fock (compute-bound with atomic operations); and compare their performance against vendor baselines on NVIDIA H100 and AMD MI300A GPUs. We show that Mojo’s performance is competitive with CUDA and HIP for memory-bound kernels, whereas gaps exist on AMD GPUs for atomic operations and for fast-math compute-bound kernels on both AMD and NVIDIA GPUs. Although the learning curve and programming requirements are still fairly low-level, Mojo can close significant gaps in the fragmented Python ecosystem in the convergence of scientific computing and AI.

Godoy, William [ORNL] (ORCID:0000000225905178)↗

CGSim: A Simulation Framework for Large Scale Distributed Computing Environment

Large-scale distributed computing infrastructures such as the Worldwide LHC Computing Grid (WLCG) require comprehensive simulation tools for evaluating performance, testing new algorithms, and optimizing resource allocation strategies. However, existing simulators suffer from limited scalability, hardwired algorithms, lack of real-time monitoring, and inability to generate datasets suitable for modern machine learning approaches. We present CGSim, a simulation framework for large-scale distributed computing environments that addresses these limitations. Built upon the validated SimGrid simulation framework, CGSim provides high-level abstractions for modeling heterogeneous grid environments while maintaining accuracy and scalability. Key features include a modular plugin mechanism for testing custom workflow scheduling and data movement policies, interactive real-time visualization dashboards, and automatic generation of event-level datasets suitable for AI-assisted performance modeling. We demonstrate CGSim’s capabilities through a comprehensive evaluation using production ATLAS PanDA workloads, showing significant calibration accuracy improvements across WLCG computing sites. Scalability experiments show near-linear scaling for multi-site simulations, with distributed workloads achieving 6 × better performance compared to single-site execution. The framework enables researchers to simulate WLCG-scale infrastructures with hundreds of sites and thousands of concurrent jobs within practical time budget constraints on commodity hardware.

Vatsavai, Sairam Sri [Brookhaven National Laborato↗

Deploying and Tracking Software with NCCS Software Provisioning

The National Center for Computational Sciences (NCCS) at Oak Ridge National Laboratory has a long history of deploying ground-breaking leadership-class supercomputers for the U.S. Department of Energy. The latest in this line of supercomputers is Frontier, the first supercomputer to break the exascale barrier (1018 floating-point operations per second) on the TOP500 list. Frontier serves a wide array of scientific domains, from traditional simulation-based workloads to newer AI and Machine Learning workloads. To best serve the NCCS user community, NCCS uses Spack to deploy a comprehensive software stack of scientific software packages, providing straightforward access to these packages through Lmod Environment Modules. Maintaining a large software stack while also including multiple new compiler releases each year is a very time-consuming task. Additionally, it is not straightforward to provide a software stack alongside existing vendor-provided software such as the HPE/Cray Programming Environment (CPE), and existing CPE, Spack, and Lmod integration does not allow for multiple versions of GPU libraries such as AMD’s ROCm to be used. To address these challenges and shortcomings, NCCS has developed the NCCS Software Provisioning tool (NSP)1, a tool for deploying and monitoring software stacks on HPC systems. NSP allows NCCS to quickly and effectively provision software stacks from the ground up using template-driven recipes and configuration files. NSP is successfully deployed on Frontier and several other NCCS clusters, enabling the NCCS software team to quickly deploy software stacks for newly-released compilers, expand current software offerings, better support GPU-based software, and monitor Lmod module usage to identify unused software packages that can be removed from the software stack. In this work, we discuss the shortcomings of the previous CPE, Spack, and Lmod usage at NCCS, provide further details on the implementation and structure of NSP, then discuss the benefits that NSP provides.

Rentschler, Asa [ORNL] (ORCID:0009000597694743)↗

Q-IRIS: The Evolution of the IRIS Task-Based Runtime to Enable Classical-Quantum Workflows

Extreme heterogeneity in emerging HPC systems are starting to include quantum accelerators, motivating runtimes that can coordinate between classical and quantum workloads. We present a proof-of-concept hybrid execution framework integrating the IRIS asynchronous task-based runtime with the XACC quantum programming framework via the Quantum Intermediate Representation Execution Engine (QIR-EE). IRIS orchestrates multiple programs written in the quantum intermediate representation (QIR) across heterogeneous backends (including multiple quantum simulators), enabling concurrent execution of classical and quantum tasks. Although not a performance study, we report measurable outcomes through the successful asynchronous scheduling and execution of multiple quantum workloads. To illustrate practical runtime implications, we decompose a four-qubit circuit into smaller subcircuits through a process known as quantum circuit cutting, reducing per-task quantum simulation load and demonstrating how task granularity can improve simulator throughput and reduce queueing behavior -- effects directly relevant to early quantum hardware environments. We conclude by outlining key challenges for scaling hybrid runtimes, including coordinated scheduling, classical-quantum interaction management, and support for diverse backend resources in heterogeneous systems.

Miniskar, Narasinga Rao [ORNL] (ORCID:000000018259↗

Synthetic Proxy Infrastructure for Task Evaluation

The Synthetic Proxy Infrastructure for Task Evaluation is a proxy application designed to support application developers in gauging the performance of various task granularities when determining how best to utilize task based programming models.The infrastructure is designed to provide examples of common communication patterns with a synthetic workload intended to provide performance data to evaluate programming model and platform overheads for the purpose of determining task granularity for task decomposition purposes. This is presented as a reference implementation of a proxy application with run-time configurable input and output task dependencies ranging from an embarrassingly parallel scenario to patterns with stencil-like dependencies upon their nearest neighbors. Once all, if any, inputs are satisfied each task will execute a synthetic workload (a simple DGEMM of in this case) of varying size and output all, if any, outputs to the next tasks.The intent is for this reference implementation to be implemented as a proxy app in different programming models so as to provide the same infrastructure and to allow for application developers to simulate their own communication needs to assist in task decomposition under various models on a given platform.

Junghans, Christoph↗

Flux wrapper scripts

The Flux Wrapper Scripts translate the commands used by other HPC resource / workload management tools to Flux commands in order to provide a bridge for HPC users between other HPC resource / workload management tools and the Flux resource manager.

Day, Ryan↗