Search NASA⌕ Search

SEARCH · Search NASA

Results for “High performance computing (HPC)”

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 1,369 records · Page 76

Automatic Generation of High-Performance Convolution Kernels on ARM CPUs for Deep Learning

In this work, we present FastConv, a template-based code auto-generation open source library that can automatically generate high-performance deep learning convolution kernels of arbitrary matrices/tensors shapes. FastConv is based on the Winograd algorithm, which is reportedly the highest performing algorithm for the time-consuming convolution layers of convolutional neural networks. ARM CPUs cover a wide range designs and specifications, from embedded devices to HPC-grade CPUs. The leads to the dilemma of how to consistently optimize Winograd-based convolution solvers for convolution layers of different shapes. FastConv addresses this problem by using templates to auto-generate multiple shapes of tuned kernels variants suitable for skinny tall matrices. As a performance portable library, FastConv transparently searches for the best combination of kernel shapes, cache tiles, scheduling of loop orders, packing strategies, access patterns, and online/offline computations. Auto-tuning is used to search the parameter configuration space for the best performance for a given target architecture and problem size. The experiments with layer-wise evaluation on the VGG--16 model confirms a 1.25x performance gains is got by tuning the Winograd library. Integrated comparison results shows 1.02x to 1.40x, 1.14x to 2.17x, and 1.22x and 2.48x speedup is achieved over NNPACK, Arm NN, and FeatherCNN on the Kunpeng 920 beside few cases. Furthermore, problem size performance portability experiments with various convolution shapes shows that FastConv achieves 1.2x to 1.7x speedup and 2x to 22x speedup over NNPACK and ARM NN inference engine using Winograd on Kunpeng 920 . CPU performance portability evaluation on the VGG--16 show an average speedup over NNPACK of 1.42x, 1.21x, 1.26x, 1.37x, 2.26x, and 11.02x is observed on Kunpeng 920, Snapdragon 835, 855, 888, Apple M1, and AWS Graviton2, respectively.

97 MATHEMATICS AND COMPUTING↗

FitCache: A Transparent Drop-In Framework for Multi-Tier Caching to Accelerate Distributed Deep Learning Workloads

Training in Deep learning (DL) remains highly compute- and data-intensive, with I/O becoming a critical bottleneck as models and datasets scale. Recent studies report that data loading can dominate training time, especially on large-scale HPC systems with shared parallel file systems (PFS). Existing caching approaches either rely on single-tier designs or require intrusive modifications to training pipelines, limiting their portability and effectiveness. In this work, we present FitCache, a transparent drop-in framework for multi-tier caching to accelerate distributed DL training by coordinating fast local memory (e.g., DRAM, Persistent Memory (PMem)) and NVMe as hierarchical caches atop PFS. Our design adapts to hardware diversity, i.e., if NVMe is missing, memory transparently acts as a caching tier, ensuring stable performance. FitCache transparently intercepts I/O requests and issues concurrent fetches across all tiers, returning data from the fastest responder without centralized metadata or static redirection paths. FitCache adapts to dynamic workloads and heterogeneous clusters while maintaining POSIX compatibility. Experiments on Frontier (2048 GPUs) and smaller research clusters show that FitCache reduces training time by up to 40% and per-batch I/O latency by up to 71.6% compared to Lustre Orion PFS, offering a drop-in solution for scalable DL training.

Hu, Guangxing [ORNL] (ORCID:0009000283203614)↗

High-Throughput Computing: Case Study of Medical Image Processing Applications

HPC is designed for large-scale simulations using monolithic codes of tightly coupled processes highly optimized to deliver decreased time to solution. Medical image processing is not a traditional field of HPC. Similar to AI applications, medical image processing parses large datasets, typically multiple times, to support a variety of studies for classification, diagnosis or monitoring purposes. The convergence of AI, HPC and Big Data encouraged more fields using image processing to transition to HPC. However, not all applications benefit from the same optimizations. In this paper we focus on high throughput medical image processing applications that analyze a huge dataset of small MRI images and that require HPC systems to decrease the time of parsing the entire dataset and not individual MRIs. We show in this research the performance of running SLANT, an image processing application for a whole brain segmentation, on large-scale systems and highlight performance limitations. We present optimizations prioritizing throughput that exhibit a 3.5x speed-up on the Summit Supercomputer that can be used as a baseline for building a high-throughput execution framework for other HPC systems.

Predescu, Maria↗

HPC Analytics Support. Requirements for Uncertainty Quantification Benchmarks

This report outlines techniques for extending benchmark generation products so they support uncertainty quantification by benchmarked systems. We describe how uncertainty quantification requirements can be presented to candidate analytical tools supporting SPARQL. We describe benchmark data sets for evaluating uncertainty quantification, as well as an approach for using our benchmark generator to produce data sets for generating benchmark data sets.

benchmark↗

A High-Performance Computing GNSS-aware Path Planning Algorithm for Safe Urban Flight Operations

The emergence and development of advanced technologies and vehicle types have created a growing demand for new forms of flight operations. These new and increasingly complex operational paradigms, such as Advanced and Urban Air Mobility (AAM/UAM), present regulatory authorities and the aviation community with several design-and-implementation challenges – particularly for highly autonomous vehicles. An overarching and daunting task is to develop protocols that can integrate these operations without compromising safety or disrupting traditional airspace operations. A shift toward a more predictive, autonomous, risk mitigation capability becomes critical to meet this challenge. This paper proposes and evaluates a computationally-efficient path planning approach to perform pre-flight planning and autonomous in-flight re-routing to minimize exposures to selected hazards. In our evaluation, hazards associated with degraded and missing critical GPS navigation data are considered. In this paper, we first present a high-performance computing path planning approach based on an adapted Bellman-Ford algorithm, developed in the CUDA programming language. Using the adapted path planning algorithm, we test this algorithm when encountering issues with GPS quality, and deliver an implementation that can produce flight paths that minimize exposure to risks, while maintaining a low computational burden. In our evaluation, the computation of periodic and aperiodic path updates are evaluated, prioritizing specific events as triggers for updates, based on changes to satellite availability. These critical events can lead to significant exposure to navigational hazards if not dealt with correctly.

GNSS↗

A High-Performance Computing GNSS-aware Path Planning Algorithm for Safe Urban Flight Operations

The emergence and development of advanced technologies and vehicle types have created a growing demand for new forms of flight operations. These new and increasingly complex operational paradigms, such as Advanced and Urban Air Mobility (AAM/UAM), present regulatory authorities and the aviation community with several design-and-implementation challenges – particularly for highly autonomous vehicles. An overarching and daunting task is to develop protocols that can integrate these operations without compromising safety or disrupting traditional airspace operations. A shift toward a more predictive, autonomous, risk mitigation capability becomes critical to meet this challenge. This paper proposes and evaluates a computationally-efficient path planning approach to perform pre-flight planning and autonomous in-flight re-routing to minimize exposures to selected hazards. In our evaluation, hazards associated with degraded and missing critical GPS navigation data are considered. In this paper, we first present a high-performance computing path planning approach based on an adapted Bellman-Ford algorithm, developed in the CUDA programming language. Using the adapted path planning algorithm, we test this algorithm when encountering issues with GPS quality, and deliver an implementation that can produce flight paths that minimize exposure to risks, while maintaining a low computational burden. In our evaluation, the computation of periodic and aperiodic path updates are evaluated, prioritizing specific events as triggers for updates, based on changes to satellite availability. These critical events can lead to significant exposure to navigational hazards if not dealt with correctly.

GNSS↗

A Methodology for the Rapid Development of Scalable HPC Data Services

There is growing evidence in the scientifc computing community that parallel file systems are not sufficient for all HPC storage workloads. This realization has motivated extensive research in new storage system designs. The question of which alternative we should turn towards implies that there could be a single answer satisfying a wide range of very diverse applications. We argue that such a generic solution does not exist. Instead, custom data services should be designed and tailored to the needs of specific applications on specific hardware. Furthermore, custom data services should be designed in close collaboration with users. In this paper, we present a methodology for the rapid development of such data services. This methodology promotes the design of reusable building blocks that can be composed together efficiently through a runtime based on high-performance threading, tasking, and remote procedure calls. We illustrate the success of our methodology by showcasing three examples of data services designed from the same building blocks, yet targeting entirely different applications.

Dorier, Mattieu↗

Harnessing the power of supercomputers using the PanDA Pilot 2 in the ATLAS Experiment

The unprecedented computing resource needs of the ATLAS experiment at LHC have motivated the Collaboration to become a leader in exploiting High Performance Computers (HPCs). To meet the requirements of HPCs, the PanDA system has been equipped with two new components; Pilot 2 and Harvester, that were designed with HPCs in mind. While Harvester is a resource-facing service which provides resource provisioning and workload shaping, Pilot 2 is responsible for payload execution on the resource. The presentation focuses on Pilot 2, which is a complete rewrite of the original PanDA Pilot used by ATLAS and other experiments for well over a decade. Pilot 2 has a flexible and adaptive design that allows for plugins to be defined with streamlined workflows. In particular, it has plugins for specific hardware infrastructures (HPC/GPU clusters) as well as for dedicated workflows defined by the needs of an experiment. Examples of dedicated HPC workflows are discussed in which the Pilot either uses an MPI application for processing fine-grained event level service under the control of the Harvester service or acts like an MPI application itself and runs a set of job in an assemble. In addition to describing the technical details of these workflows, results are shown from its deployment on Titan (OLCF) and other HPCs in ATLAS.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

W17_geowave “3D full waveform geophysical models”

Performance of the MCMC inversion according to the number of cores for the computation. A) 64 cores. B) 480 cores. C) 816 cores. The true model is represented by the black line. Vsv is the wave speed of S waves polarized in the vertical plane, ξ is an anisotropy parameter. The Earth is highly anisotropics; the wavespeed of seismic waves depends on the polarization of the wave. Seismic inversion of the elastic structure is usually limited to isotropic information such as Vsv. Our research looked at the inversion of Earth anisotropy.

58 GEOSCIENCES↗

SU-E-T-47: Application of the Repair-Misrepair-Fixation RBE Model to Describe the Results of High Resolution Proton Irradiation Cell Survival Experiments

Purpose: To develop a system to rapidly and accurately calculate RBE with the repair-misrepair-fixation (RMF) model for proton therapy data sets and to evaluate its effectiveness in modeling RBE for cell survival experiments performed with the H460 cell line for a range of proton LET. Methods: A system for using the Monte Carlo Damage Simulation (MCDS) software with high performance computing was developed. Input for the MCDS software for a range of proton energies in increments of 0.1 MeV was generated and simulated. The output data were then used to determine doseaveraged quantities for the RMF model based on equivalently binned proton energy spectra. The method was applied to calculate RBE at 50% survival for experimental cell survival data. Experimental data were obtained using a system which allowed for the accumulation of cell survival data at known values of dose-averaged proton LETs at a range of doses. RBE was calculated based directly on a Cs-137 reference experiment and, additionally, according to fitted values of the θ and κ terms of the RMF model. Results: Dose-averaged RMF model quantities were calculated using the HPC system. Compared to experimental RBE determined using a Cs-137 irradiation as a reference, the RBE from the model differed by at most 49%. RBE based on the fitted values of θ and κ differed by at most 18% for the highest LET. Conclusion: A system for rapidly generating data necessary to calculate RBE with the RMF model has been developed. For the H460 cell line, the RMF model could not reproduce the experimentally determined RBE based solely on the photon reference data. Fitting of the θ and κ terms of the RMF model indicates that their values increase for proton LET exceeding approximately 10 keV/µm. NIH Program Project Grant P01CA021239.

60 APPLIED LIFE SCIENCES↗

Kinetic Monte Carlo Simulations of Chemical Reactions on Surfaces

Lattice based kinetic Monte Carlo (kMC) simulations are a powerful tool for micro-kinetic modeling. The kinetic Monte Carlo technique stochastically simulates chemical reaction processes (e.g. adsorption, desorption, diffusion), where the probability of a given reaction process being selected is weighted by its rate constant. The open source software package, KMOS1, allows fast construction of kMC simulations by automatically implementing a computationally efficient kMC framework, thus allowing the researcher to focus on accurate parameterization and formulation of the kinetic model. From this, the spatiotemporal evolution (e.g. concentration, surface configuration, turnover frequencies) of crystal surfaces or bulk crystalline structures is obtained. KMOS uses the variable step-size method implementation of kMC. At each step, the VSSM updates available events, calculates the corresponding transition rates and simulation time, and selects one of the available events. The lattice based kMC technique has many potential applications within SRNL where chemical kinetics are of interest. By transitioning through time along a trajectory of configuration space, fundamental insights to a system's dominant reactions and parameters can be gained, leading to improved materials and process design. In this work, we have established capabilities for fast implementation of kMC simulations through the installation and testing of the KMOS software on SRNL high performance computing resources. The parameterization of kMC simulations is often performed from first-principles energetic calculations. For the purposes of this work, the test cases use arbitrary species and thus, have arbitrary parameterization. In order to demonstrate the use of KMOS, two primary test cases are shown. Case 1 involves simple adsorption and desorption. Case 2 involves a slightly more complex mechanism. The evolution of the surface concentration and the final surface configuration for simulation Case 2 is shown. Rate-limiting processes and rate-limiting configurations are not always known or understood, but kMC provides information pertaining to both. The capability to implement efficient micro-kinetic models using the kinetic Monte Carlo technique has been established through the testing and installation of KMOS on SRNL HPC resources.

97 MATHEMATICS AND COMPUTING↗

OpenSHMEM-UCX: Evaluation of UCX for Implementing OpenSHMEM Programming Model, In: OpenSHMEM 2016: OpenSHMEM and Related Technologies. Enhancing OpenSHMEM for Hybrid Environments

The OpenSHMEM reference implementation was developed towards the goal of developing an open source and high-performing OpenSHMEM implementation. To achieve portability and performance across various networks, the OpenSHMEM reference implementation uses GASNet and UCCS for network operations. Recently, new network layers have emerged with the promise of providing high-performance, scalability, and portability for HPC applications. In this paper, we implement the OpenSHMEM reference implementation to use the UCX framework for network operations. Then, we evaluate its performance and scalability on Cray XK systems to understand UCX’s suitability for developing the OpenSHMEM programming model. Further, we develop a benchmark called SHOMS for evaluating the OpenSHMEM implementation. Our experimental results show that OpenSHMEM-UCX outperforms the vendor supplied OpenSHMEM implementation in most cases on the Cray XK system by up to 40% with respect to message rate and up to 70% for the execution of application kernels.

Computer Science↗

HPC Programming on Intel Many-Integrated-Core Hardware with MAGMA Port to Xeon Phi

This paper presents the design and implementation of several fundamental dense linear algebra (DLA) algorithms for multicore with Intel Xeon Phi coprocessors. In particular, we consider algorithms for solving linear systems. Further, we give an overview of the MAGMA MIC library, an open source, high performance library, that incorporates the developments presented here and, more broadly, provides the DLA functionality equivalent to that of the popular LAPACK library while targeting heterogeneous architectures that feature a mix of multicore CPUs and coprocessors. The LAPACK-compliance simplifies the use of the MAGMA MIC library in applications, while providing them with portably performant DLA. High performance is obtained through the use of the high-performance BLAS, hardware-specific tuning, and a hybridization methodology whereby we split the algorithm into computational tasks of various granularities. Execution of those tasks is properly scheduled over the heterogeneous hardware by minimizing data movements and mapping algorithmic requirements to the architectural strengths of the various heterogeneous hardware components. Our methodology and programming techniques are incorporated into the MAGMA MIC API, which abstracts the application developer from the specifics of the Xeon Phi architecture and is therefore applicable to algorithms beyond the scope of DLA.

97 MATHEMATICS AND COMPUTING↗

Integrating Distributed-Memory Machine Learning into Large-Scale HPC Simulations

Lawrence Livermore National Laboratory (LLNL) is a national lab based in Livermore, California. It was established in 1952 to meet increasing national security needs. They specialize in earth and atmospheric science, bioscience and bioengineering, lasers and optical science and technology, advanced materials and manufacturing, high-energy-density science, nuclear, chemical, and isotopic science and technology, and HPC, simulation and data science ((cor, 2018)). LLNL is also home to a number of large Linux clusters and supercomputers which are used to support research in their areas of specialization (mac, 2018). One purpose of the LLNL supercomputers is to run highly detailed and accurate physics simulations. These simulations allow scientists to get data on experiments that would be too expensive, difficult, or dangerous to perform in reality. With the help of simulations scientists at LLNL are able to study processes in more depth than they would be able to otherwise.

97 MATHEMATICS AND COMPUTING↗

Toward an optimal online checkpoint solution under a two-level HPC checkpoint model

The traditional single-level checkpointing method suffers from significant overhead on large-scale platforms. Hence, multilevel checkpointing protocols have been studied extensively in recent years. The multilevel checkpoint approach allows different levels of checkpoints to be set (each with different checkpoint overheads and recovery abilities), in order to further improve the fault tolerance performance of extreme-scale HPC applications. How to optimize the checkpoint intervals for each level, however, is an extremely difficult problem. In this paper, we construct an easy-to-use two-level checkpoint model. Checkpoint level 1 deals with errors with low checkpoint/recovery overheads such as transient memory errors, while checkpoint level 2 deals with hardware crashes such as node failures. Compared with previous optimization work, our new optimal checkpoint solution offers two improvements: (1) it is an online solution without requiring knowledge of the job length in advance, and (2) it shows that periodic patterns are optimal and determines the best pattern. We evaluate the proposed solution and compare it with the most up-to-date related approaches on an extreme-scale simulation testbed constructed based on a real HPC application execution. Simulation results show that our proposed solution outperforms other optimized solutions and can improve the performance significantly in some cases. Specifically, with the new solution the wall-clock time can be reduced by up to 25.3% over that of other state-of-the-art approaches. Lastly, a brute-force comparison with all possible patterns shows that our solution is always within 1% of the best pattern in the experiments.

97 MATHEMATICS AND COMPUTING↗

HPC application fault-tolerance using transparent redundant computation.

As the core count of HPC machines continue to grow in size, issues such as fault tolerance and reliability are becoming limiting factors for application scalability. Current techniques to ensure progress across faults, for example coordinated checkpoint-restart, are unsuitable for machines of this scale due to their predicted high overheads. In this study, we present the design and implementation of a novel system for ensuring reliability which uses transparent, rank-level, redundant computation. Using this system, we show the overheads involved in redundant computation for a number of real-world HPC applications. Additionally, we relate the communication characteristics of an application to the overheads observed.

97 MATHEMATICS AND COMPUTING↗

Dynamic Undervolting to Improve Energy Efficiency on Multicore X86 CPUs

Chip manufacturers introduce redundancy at various levels of CPU design to guarantee correct operation, even for worst-case combinations of non-idealities in process variation and system operating conditions. This redundancy is implemented partly in the form of voltage margins. However, for a wide range of real-world execution scenarios these margins are excessive and merely translate to increased power consumption, hindering the effort towards higher-energy efficiency in both HPC and general purpose computing. Our study on the x86-64 Haswell and Skylake multicore microarchitectures reveals-wide voltage margins, which vary across different microarchitectures, different chip parts of the same microarchitecture, and across different workloads. We find that it is necessary to quantify-voltage margins using multi-threaded and multi-instance workloads, as characterization with single-threaded and single-instance workloads that do not stress the CPU to its full capacity typically identifies overly optimistic margins that lead to errors when applied in realistic program execution scenarios. In addition, we introduce, deploy and evaluate a run-time governor that dynamically reduces the supply voltage of modern multicore x86-64 CPUs. Our governor employs a model that takes as input a set of performance metrics which are directly measurable via performance monitoring counters and have high predictive value for the minimum tolerable supply voltage (V min ), to predict and apply the appropriate reduction for the workload at hand. Compared with the conventional DVFS governor, our approach in this study achieves up to 42 percent energy savings for the Skylake family and 34 percent for the Haswell family for complex, real-world applications.

97 MATHEMATICS AND COMPUTING↗

Continuous whole-system monitoring toward rapid understanding of production HPC applications and systems

A detailed understanding of HPC applications’ resource needs and their complex interactions with each other and HPC platform resources are critical to achieving scalability and performance. Such understanding has been difficult to achieve because typical application profiling tools do not capture the behaviors of codes under the potentially wide spectrum of actual production conditions and because typical monitoring tools do not capture system resource usage information with high enough fidelity to gain sufficient insight into application performance and demands. In this paper we present both system and application profiling results based on data obtained through synchronized system wide monitoring on a production HPC cluster at Sandia National Laboratories (SNL). We demonstrate analytic and visualization techniques that we are using to characterize application and system resource usage under production conditions for better understanding of application resource needs. Furthermore, our goals are to improve application performance (through understanding application-to-resource mapping and system throughput) and to ensure that future system capabilities match their intended workloads.

97 MATHEMATICS AND COMPUTING↗