SEARCH · Search NASA
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.
This two-page fact sheet describes the new High Performance Computing Data Center in the ESIF and talks about some of the capabilities and unique features of the center.
With the growing scale and complexity of high-performance computing (HPC) systems, resilience solutions that ensure continuity of service despite frequent errors and component failures must be methodically designed to balance the reliability requirements with the overheads to performance and power. Design patterns enable a structured approach to the development of resilience solutions, providing hardware and software designers with the building block elements for the rapid development of novel solutions and for adapting existing technologies for emerging, extreme-scale HPC environments. In this paper, we develop analytical models that enable designers to evaluate the reliability and performance characteristics of the design patterns. These models are particularly useful in building a unified framework that analyzes and compares various resilience solutions built using a combination of patterns.
In March of 2020, recognizing the potential of High Performance Computing (HPC) to accelerate understanding and the pace of scientific discovery in the fight to stop COVID-19, the HPC community assembled the largest collection of worldwide HPC resources to enable COVID-19 researchers worldwide to advance their critical efforts. Amazingly, the COVID-19 HPC Consortium was formed within one week through the joint effort of the Office of Science and Technology Policy (OSTP), the U.S. Department of Energy (DOE), the National Science Foundation (NSF), and IBM to create a unique public–private partnership between government, industry, and academic leaders. This article is the Consortium's story–how the Consortium was created, its founding members, what it provides, how it works, and its accomplishments. We will reflect on the lessons learned from the creation and operation of the Consortium and describe how the features of the Consortium could be sustained as a National Strategic Computing Reserve to ensure the nation is prepared for future crises.
Resiliency is the ability of large-scale high-performance computing (HPC) applications to gracefully handle errors, and recover from failures. In this paper, we propose a pattern-based approach to constructing resilience solutions that handle multiple error modes. Using resilience patterns, we evaluate the performance and reliability characteristics of detection, containment and mitigation techniques for transient errors that cause silent data corruptions and techniques for fail-stop errors that result in process failures. We demonstrate the design and implementation of the multiresilience solution based on patterns instantiated across multiple layers of the system stack. The patterns are integrated to work together to achieve resiliency to different error types in a performance-efficient manner.
Resource allocation in High Performance Computing (HPC) environments presents a complex and multifaceted challenge for job scheduling algorithms. Beyond the efficient allocation of system resources, schedulers must account for and optimize multiple performance metrics, including job wait time and system throughput. Traditional heuristic-based scheduling algorithms increasingly struggle and lack the efficiency needed to meet the demands and address the complexity and scale of modern HPC systems. Consequently, recent research efforts have focused on leveraging advancements in Artificial Intelligence (AI) and Deep Learning (DL), particularly Reinforcement Learning (RL), to develop more adaptable and intelligent scheduling strategies. Previous RL-based scheduling approaches have explored a range of algorithms, from Deep Q-Networks (DQN) to Proximal Policy Optimization (PPO), and more recently, hybrid methods that integrate Graph Neural Networks (GNNs) with RL techniques. However, a common limitation across these methods is their reliance on relatively small datasets, with few methods being evaluated using large-scale, multi-million-job trace datasets representative of real-world HPC workloads. Moreover, existing RL schedulers face scalability issues due to centralized policy updates, which hinder training efficiency and performance when applied to large datasets. This study introduces a novel RL-based scheduler utilizing Decentralized Distributed Proximal Policy Optimization (DD-PPO) algorithm, which supports large-scale distributed training across multiple workers without requiring parameter synchronization at every step. By eliminating reliance on centralized updates to a shared policy, the DD-PPO scheduler enhances scalability, training efficiency, and sample utilization. Experimental validation using a large real-world dataset containing over 11.5 million job traces collected from petascale HPC systems over six years assesses the influence of dataset scale on training effectiveness and compares DD-PPO performance to traditional and advanced scheduling approaches. The experimental results demonstrate improved scheduling performance in comparison to both heuristic-based schedulers and existing RL-based scheduling algorithms.
The dataset comprises trace files from high performance computing (HPC) simulations. The trace files contain records of every I/O operation executed by a simulation application run, including I/O operations from HDF5, MPI-IO, and POSIX and all of the parameters supplied to those operations, e.g. file name, offset, and flags. The traces are generated by executing a simulation application that is linked with the Recorder tracing tool (https://github.com/uiuc-hpc/Recorder). The Recorder trace tool intercepts the I/O calls made by the application, records the I/O trace record, and then calls the intended I/O call so that the operation executes.
The demand for high performance computing (HPC) resources continues to grow, driven by the increasing complexity of modeling and simulation, artificial intelligence (AI), and machine learning (ML) workloads [Porter]. The growing energy consumption demand of these HPC systems is a significant concern, both in terms of operational costs and environmental impact. AI hardware accelerators are expected to reach 1.5% of the world’s power consumption by 2029 [Shah].
In 2019, highway congestion wasted over 3 billion gallons of fuel and caused 8.8 billion hours of lost productivity.1 Research has shown that introducing near-real time traffic controls can significantly reduce congestion. Validated and calibrated traffic simulations enable the modeling of transportation systems and the evaluation of different traffic control actions and schemes given a variety of circumstances that represent likely future scenarios. The developed scenarios can inform the deployment of controls in near real-time to improve freight and passenger vehicle congestion and energy use. In this work, we present simulations used to model the traffic in the Chattanooga, Tennessee, metropolitan area. Simulations were constructed and calibrated using a variety of local, data science enhanced, data sources utilizing open source software including the Simulation of Urban Mobility (SUMO) simulator. High-Performance Computing (HPC) provides a scalable platform with enough computing for the high-fidelity simulation of many scenarios and the application of advance data science especially for large-scale systems. Our simulations include microscopic simulations at a corridor level for traffic signal control, and mesoscopic simulations to evaluate regional operational controls and infrastructure.
A PowerPoint on Enhancing High Performance Computing (HPC) User Support with an AI Assistant through a Science Gateway.
The Argonne Training Program in Extreme Scale Computing (ATPESC) was started by Argonne National Laboratory with the objective of expanding the ranks of better-prepared users of high-performance computing (HPC) machines. One of the unique aspects of the program was inclusion of a track on software engineering and community codes. The inclusion was motivated by the observation that the projects with good software processes were better able to meet their scientific goals. Over the years, with greater awareness of software sustainability issues in the community, the track has evolved into a software productivity and sustainability track. In this paper we present our experience in choosing and disseminating the content related to the topic of software engineering in high performance computing science from the beginning of the program until now. We discuss the motivations and the reception of the tracks. We also document the evolution of the track over the years based on student feedback and also the growth of awareness about software productivity in high performance computing.
Omega version 0, the first step toward a new ocean model, has been developed to take full advantage of next-generation supercomputers, including those powered by GPUs. Omega is rebuilt from the ground up, redesigning computational aspects of ocean models to run faster and more efficiently across diverse high-performance computing systems.
Traditional Lagrangian spray modeling approaches for internal combustion engines are highly griddependent due to insufficient resolution in the near nozzle region. This is primarily because of inherent restrictions of volume fraction with the Lagrangian assumption together with high computational costs associated with small grid sizes. A state-of-the-art grid-convergent spray modeling approach was developed and implemented by Senecal et al. (ASME-ICEF2012-92043) in the CONVERGE software. The key features of the methodology include Adaptive Mesh Refinement (AMR), advanced liquid-gas momentum coupling, and improved distribution of the liquid phase, which enables use of cell sizes smaller than the nozzle diameter. This modeling approach was rigorously validated against nonevaporating, evaporating, and reacting data from the literature. The current numerical study focusses on further demonstration of the grid-convergent modeling approach for simulating a single-cylinder Cat® compression ignition engine. The simulated injector is characterized with a nominal nozzle exit diameter of 259 μm. Simulations using various minimum grid sizes (ranging from 125 μm to 1000 μm) are compared for engine performance and emissions parameters of interest such as pressure, heat release rate, ignition delay, NOx, HC, and soot emissions. The peak cell-count for the highest resolution simulation was on the order of 34 million. These computationally expensive simulations were facilitated at a high-performance computing facility at Argonne National Laboratory. METIS load-balancing algorithm was developed and implemented in Converge code for the simulations. Scaling studies were also performed. The validity of previously recommended grid settings (ASME-ICEF2012-92043) for accuracy/runtime trade-off is further assessed. Efficacy of a simplified combustion model is also compared against a detailed chemical kinetics based combustion model.
This report highlights a subset of accomplishments including projects that were completed in FY 2019 by researchers using Idaho National Laboratory’s High-Performance Computing (HPC) resources.
Federated learning enables multiple data owners to collaboratively train robust machine learning models without transferring large or sensitive local datasets by only sharing the parameters of the locally trained models. Here, in this article, we elaborate on the design of our Advanced Privacy-Preserving Federated Learning (APPFL) framework, which streamlines end-to-end secure and reliable federated learning experiments across cloud computing facilities and high-performance computing resources by leveraging Globus Compute, a distributed function as a service platform, and Amazon Web Services. We further demonstrate the use case of APPFL in fine-tuning an LLaMA 2 7B model using several cloud resources and supercomputers.
This is a set of slides from a guest lecture for a class at the University of Texas, El Paso on visualization and data analysis for high-performance computing. The topics covered are the following: trends in high-performance computing; scientific visualization, such as OpenGL, ray tracing and volume rendering, VTK, and ParaView; data science at scale, such as in-situ visualization, image databases, distributed memory parallelism, shared memory parallelism, VTK-m, "big data", and then an analysis example.
Accelerator devices are becoming a norm in High Performance Computing (HPC). With more systems opting for heterogeneous architectures, portable programming models like OpenMP and OpenACC are becoming increasingly important. The SPEC ACCEL 1.2 benchmark suite consists of comparable benchmarks in OpenCL, OpenMP 4.5, and OpenACC 2.5 that can be used to evaluate the performance and support for programming models and frameworks on heterogeneous platforms. In this paper we go beneath the normative metric of performance times and look at the individual kernels to study the usage, strengths, and weaknesses of the two prevalent portable heterogeneous programming models, OpenMP and OpenACC. From our analysis we identify that benchmarks like MRI-Q, SP and BT have better performance using OpenACC, while benchmarks like MiniGhost, LBM and LBDC do consistently better with the OpenMP programming model across supercomputers like Titan, and Summit.We deep dive into the kernels of select four benchmarks to answer questions like: Where does the benchmark spend most of its cycles? What is the parallelization strategy used? Why is one programming model more performant than the other? By identifying the similarities and dierences we want to contrast between the benchmark implementation strategies in the SPEC ACCEL 1.2 benchmarks and provide more insights into the OpenMP and OpenACC programming models.
There are multiple nuclear microreactors currently under development that are designed to provide autonomous power for as many as ten or more years without refueling and are designed to power high performance computing (HPC) datacenters. But the load-follow speeds for a nuclear microreactor will be much slower than grid power and slower than the power variance typical of a HPC system. HPC datacenters experience peak power load variance driven by several factors ranging from the operation of cooling systems to remove heat from the servers to supporting a wide range of user application workflows and architectures each with different power signatures. One mechanism to support the limited load-follow of a microreactor is peak shaving where an energy storage mechanism is used to shed peak load and reduce significant power variance. This work explores peak electrical load shaving using uninterruptible power supply (UPS) systems designed for HPC support in the context of peak shaving when operating using a nuclear microreactor with a load-follow limited to 10% of load per minute. Using a self contained HPC datacenter complete with stand-alone cooling system and provisioned with an x86 cluster, an ARM cluster, and a graphics processing unit (GPU) cluster, peak shaving for microreactor operation using the UPS battery backup is explored while running two classes of typical HPC user applications. HPC architecture suitability for microreactor operation under this type of peak shaving is examined.
Ensuring the reliability of applications is becoming an increasingly important challenge as high-performance computing (HPC) systems experience an ever-growing number of faults, errors and failures. While the HPC community has made substantial progress in developing various resilience solutions, it continues to rely on platform-based metrics to quantify application resiliency improvements. The resilience of an HPC application is concerned with the reliability of the application outcome as well as the fault handling efficiency. To understand the scope of impact, effective coverage and performance efficiency of existing and emerging resilience solutions, there is a need for new metrics. In this paper, we develop new ways to quantify resilience that consider both the reliability and the performance characteristics of the solutions from the perspective of HPC applications. As HPC systems continue to evolve in terms of scale and complexity, it is expected that applications will experience various types of faults, errors and failures, which will require applications to apply multiple resilience solutions across the system stack. The proposed metrics are intended to be useful for understanding the combined impact of these solutions on an application's ability to produce correct results and to evaluate their overall impact on an application's performance in the presence of various modes of faults.