Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “HPC PROBLEMS”

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 199 records · Page 11

Benchmarking Initiatives

Benchmarking has been recognized as an efficient tool for model validation as well as scientific collaboration. Several workshop series have been established to foster the benchmarking idea: setting up test cases with increasing complexity for method development and code comparison. In addition to representing model complexity, one of the key efforts is to develop codes that are suitable for modern HPC platforms such as PetaFlop supercomputers. It has been realized that those challenges are beyond single team capabilities. The Subsurface Environmental Simulation Benchmarking (SSBench) working group was initiated to develop, publish, and make publicly available well-described benchmark problems that could be used to demonstrate simulator conformance with norms established by the subsurface science and engineering community. SSBench supports the development of environmentally-relevant benchmark problem sets that provide rigorous tests for process model conceptualization, numerical implementation, process coupling, and accuracy. The initial SSBench focus on multicomponent reactive transport modeling reflects the expanding range of mechanisms and increasing complexity needed to address issues relevant to climate change, nuclear waste management, contaminant remediation, and pollution prevention.

benchmarking↗

Detailed Modeling and Evaluation of a Scalable Multilevel Checkpointing System

High-performance computing (HPC) systems are growing more powerful by utilizing more components. As the system mean time before failure correspondingly drops, applications must checkpoint frequently to make progress. But, at scale, the cost of checkpointing becomes prohibitive. A solution to this problem is multilevel checkpointing, which employs multiple types of checkpoints in a single run. Moreover, lightweight checkpoints can handle the most common failure modes, while more expensive checkpoints can handle severe failures. We designed a multilevel checkpointing library, the Scalable Checkpoint/Restart (SCR) library, that writes lightweight checkpoints to node-local storage in addition to the parallel file system. We present probabilistic Markov models of SCR's performance. We show that on future large-scale systems, SCR can lead to a gain in machine efficiency of up to 35 percent, and reduce the load on the parallel file system by a factor of two. In addition, we predict that checkpoint scavenging, or only writing checkpoints to the parallel file system on application termination, can reduce the load on the parallel file system by 20 × on today's systems and still maintain high application efficiency.

97 MATHEMATICS AND COMPUTING↗

Exploring Deep Learning and Sparse Matrix Format Selection

We proposed to explore the use of Deep Neural Networks (DNN) for addressing the longstanding barriers. The recent rapid progress of DNN technology has created a large impact in many fields, which has significantly improved the prediction accuracy over traditional machine learning techniques in image classifications, speech recognitions, machine translations, and so on. To some degree, these tasks resemble the decision makings in many HPC tasks, including the aforementioned format selection for SpMV and linear solver selection. For instance, sparse matrix format selection is akin to image classification—such as, to tell whether an image contains a dog or a cat; in both problems, the right decisions are primarily determined by the spatial patterns of the elements in an input. For image classification, the patterns are of pixels, and for sparse matrix format selection, they are of non-zero elements. DNN could be naturally applied if we regard a sparse matrix as an image and the format selection or solver selection as classification problems.

97 MATHEMATICS AND COMPUTING↗

Transportation Hub Infrastructure Expansion: Decision Support Under Uncertainty

The Athena project (www.athena-mobility.org) has worked to investigate the relationship between the Dallas-Fort Worth Airport (DFW) and the greater Dallas area in order to better understand and therefore better inform future decision-making regarding the critical infrastructure that influence mobility between the airport and the city. Through this work, infrastructure related to curbside pickup and drop-off, parking, public transit, and the road network congestion were identified as critical to the operation of the DFW transportation hub. The infrastructure analysis and expansion aspect of the Athena project is focused on the restructuring of the CTA curb as a hierarchical curb and the building or repurposing of parking infrastructure as the interplay between these two areas. Many sources of uncertainty exist that may impact future airport and transportation hub operations, such as passenger volume growth, population demographic changes over time, electric vehicle (EV) adoption rates, and autonomous vehicle (AV) adoption rates. Due to these sources of uncertainty, we have selected for our research a modeling framework that can capture various types of uncertainty and hedge against those uncertainties in the optimization process. We analyze road network and curb congestion, the rise of transportation networking companies, trends in parking usage, existing policies around this infrastructure, airport revenue streams, and other contributing factors to enable infrastructure decision making with less uncertainty. To accomplish this wholistic analysis, we have developed a novel multi-stage, multi-period stochastic optimization model which considers the airport's decisions from 2025-2045 under different possible future macro trajectories and day-to-day variations in operational conditions captured as "annual representation of operations" scenarios with respective probabilities. This model has also been designed to leverage the outputs of various efforts under the Athena project to create a combined decision framework for infrastructure decisions. These various efforts include the route optimization model, the ASPIRES simulation, the mode choice model, and the SUMO traffic simulation. Our computational experiments of this system at scale have resulted in a working version of our infrastructure model which enables the explicit representation and consideration of various sources of uncertainty in the decision process to enable robust, flexible decision-making. This model has been effectively run on NREL's HPC system, Eagle, with large numbers of stochastic scenarios and shows promise as a scalable tool for robust consideration of uncertainties in airport planning. We have tested our model using 30,240 operational circumstances in total, resulting in a problem with more 200 million variables. This model was solved in several different configurations, and a workflow to simulate the performance of the infrastructure model results was developed and deployed. In general, our results indicate that a combination of remote parking, remote curb infrastructure, and dynamic pricing can generate revenue, reduce emissions, accommodate emerging technologies such as AVs and EVs, and manage airport passenger growth over time. We note the success of the proposed strategy depends on the data collection and forecasting abilities of DFW. We have also seen that the AV adoption by TNCs might necessitate larger amounts of remote curb. The results of this work inform strategies for airport infrastructure decision making, as well as demonstrate the value of an adaptable model, but also indicate that there are avenues remaining where further research would be of value.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Parallel Programming in MCNP6

Monte Carlo N-Particle (MCNP)1 is a general-purpose Monte Carlo particle transport code developed by Los Alamos National Laboratory (LANL). To efficiently handle long simulations, MCNP version 6 (MCNP6) supports parallel execution using two primary programming models: • Shared-memory task-based threading using OpenMP (Open Multi-Processing), and • Distributed-memory calculations using MPI (Message Passing Interface). The OpenMP and MPI programming models enable MCNP6 to scale from desktop systems to high-performance computing (HPC) clusters, allowing users to run MCNP in one of three parallel modes: • OpenMP-only, • MPI-only, and • Hybrid (MPI + OpenMP). The choice of parallelization mode depends on the underlying computer architecture and the characteristics of the simulation problem.

97 MATHEMATICS AND COMPUTING↗

On-demand data analytics in HPC environments at leadership computing facilities: Challenges and experiences

Scaling-up scientific data analysis and machine learning algorithms for data-driven discovery is a grand challenge that we face today. Despite the growing need for analysis from science domains that are generating `Big Data' from instruments and simulations, building high-performance analytical workflows of data-intensive algorithms have been daunting because: (i) the `Big Data' hardware and software architecture landscape is constantly evolving, (ii) newer architectures impose new programming models, and (iii) data-parallel kernels of analysis algorithms and their performance facets on different architectures are poorly understood. To address these problems, we have: (i) identified scalable data-parallel kernels of popular data analysis algorithms, (ii) implemented `Mini-Apps' of those kernels using different programming models (e.g. Map Reduce, MPI, etc.), (iii) benchmarked and validated the performance of the kernels in diverse architectures. In this paper, we discuss two of those Mini-Apps and show the execution of principal component analysis built as a workflow of the Mini-Apps. We show that Mini-Apps enable scientists to (i) write domain-specific data analysis code that scales on most HPC hardware and (ii) and offers the ability (most times with over a 10x speed-up) to analyze data sizes 100 times the size of what off-the-shelf desktop/workstations of today can handle.

Harney, John↗

HPC formulations of optimization algorithms for tensor completion

Tensor completion is a powerful tool used to estimate or recover missing values in multi-way data. It has seen great success in domains such as product recommendation and healthcare. Tensor completion is most often accomplished via low-rank sparse tensor factorization, a computationally expensive non-convex optimization problem which has only recently been studied in the context of parallel computing. In this work, we study three optimization algorithms that have been successfully applied to tensor completion: alternating least squares (ALS), stochastic gradient descent (SGD), and coordinate descent (CCD++). We explore opportunities for parallelism on shared- and distributed-memory systems and address challenges such as memory- and operation-efficiency, load balance, cache locality, and communication. Among our advancements are a communication-efficient CCD++ algorithm, an ALS algorithm rich in level-3 BLAS routines, and an SGD algorithm which combines stratification with asynchronous communication. Furthermore, we show that introducing randomization during ALS and CCD++ can accelerate convergence. We evaluate our parallel formulations on a variety of real datasets on a modern supercomputer and demonstrate speedups through 16384 cores. Further, these improvements reduce time-to-solution from hours to seconds on real-world datasets. We show that after our optimizations, ALS is advantageous on parallel systems of small-to-moderate scale, while both ALS and CCD++ provide the lowest time-to-solution on large-scale distributed systems.

97 MATHEMATICS AND COMPUTING↗

A framework for graph-based synthesis, analysis, and visualization of HPC cluster job data.

The monitoring and system analysis of high performance computing (HPC) clusters is of increasing importance to the HPC community. Analysis of HPC job data can be used to characterize system usage and diagnose and examine failure modes and their effects. This analysis is not straightforward, however, due to the complex relationships that exist between jobs. These relationships are based on a number of factors, including shared compute nodes between jobs, proximity of jobs in time, etc. Graph-based techniques represent an approach that is particularly well suited to this problem, and provide an effective technique for discovering important relationships in job queuing and execution data. The efficacy of these techniques is rooted in the use of a semantic graph as a knowledge representation tool. In a semantic graph job data, represented in a combination of numerical and textual forms, can be flexibly processed into edges, with corresponding weights, expressing relationships between jobs, nodes, users, and other relevant entities. This graph-based representation permits formal manipulation by a number of analysis algorithms. This report presents a methodology and software implementation that leverages semantic graph-based techniques for the system-level monitoring and analysis of HPC clusters based on job queuing and execution data. Ontology development and graph synthesis is discussed with respect to the domain of HPC job data. The framework developed automates the synthesis of graphs from a database of job information. It also provides a front end, enabling visualization of the synthesized graphs. Additionally, an analysis engine is incorporated that provides performance analysis, graph-based clustering, and failure prediction capabilities for HPC systems.

97 MATHEMATICS AND COMPUTING↗

Immortal rays: Rethinking random ray neutron transport on GPU architectures

The Random Ray Method (TRRM) is a recently developed adaptation of the Method of Characteristics for neutral particle transport simulations. TRRM has demonstrated excellent performance on 3D nuclear reactor benchmark problems using CPU-based compute systems. When porting to GPU-based systems, however, new performance challenges arise that are unique to processors targeting massive fine-grained parallelism. For smaller problems, or for large problems that are domain decomposed across many computational nodes, the problem size per node has insufficient parallelism to saturate GPU node resources, thus greatly limiting speedup. In this study, we report on a newly developed “immortal ray” variant of TRRM. Here, the immortal ray technique exposes significantly more fine-grained parallelism by fundamentally reformulating the numerical details of ray discretization, resulting in performance tradeoffs with significant overall benefit on GPUs. For very small 2D simulation problems we found the new immortal ray variant allowed for up to a 4.4x speedup when run on a single GPU. For larger 3D simulation problems we found the new variant improved strong scaling by 3x when run on the Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗

HPC-NMF: A High-Performance Parallel Algorithm for Nonnegative Matrix Factorization

NMF is a useful tool for many applications in different domains such as topic modeling in text mining, background separation in video analysis, and community detection in social networks. Despite its popularity in the data mining community, there is a lack of efficient distributed algorithms to solve the problem for big data sets. We propose a high-performance distributed-memory parallel algorithm that computes the factorization by iteratively solving alternating non-negative least squares (NLS) subproblems for $\WW$ and $\HH$. It maintains the data and factor matrices in memory (distributed across processors), uses MPI for interprocessor communication, and, in the dense case, provably minimizes communication costs (under mild assumptions). As opposed to previous implementation, our algorithm is also flexible: It performs well for both dense and sparse matrices, and allows the user to choose any one of the multiple algorithms for solving the updates to low rank factors $\WW$ and $\HH$ within the alternating iterations.

Kannan, Ramakrishnan↗

Large-Scale Continuous Subgraph Queries on Streams

Graph pattern matching involves finding exact or approximate matches for a query subgraph in a larger graph. It has been studied extensively and has strong applications in domains such as computer vision, computational biology, social networks, security and finance. The problem of exact graph pattern matching is often described in terms of subgraph isomorphism which is NP-complete. The exponential growth in streaming data from online social networks, news and video streams and the continual need for situational awareness motivates a solution for finding patterns in streaming updates. This is also the prime driver for the real-time analytics market. Development of incremental algorithms for graph pattern matching on streaming inputs to a continually evolving graph is a nascent area of research. Some of the challenges associated with this problem are the same as found in continuous query (CQ) evaluation on streaming databases. This paper reviews some of the representative work from the exhaustively researched field of CQ systems and identifies important semantics, constraints and architectural features that are also appropriate for HPC systems performing real-time graph analytics. For each of these features we present a brief discussion of the challenge encountered in the database realm, the approach to the solution and state their relevance in a high-performance, streaming graph processing framework.

99 GENERAL AND MISCELLANEOUS↗

HEPnOS: a Specialized Data Service for High Energy Physics Analysis

In this paper, we present HEPnOS, a distributed data service for managing data produced by high-energy physics (HEP) experiments. Using HEPnOS, HEP applications can use HPC resources more effciently than traditional fle-based applications. The fle-based model leads to a rigid, chunk-based allocation of computational resources and limits the number of cores that can be used concurrently by an HEP application. The fundamental problem is that organizing domain-specifc data into fles inadvertently introduces a single, artifcial, confated tuning parameter that puts key optimization goals into confict: larger fle sizes reduce metadata overhead and thus improve I/O effciency, but smaller fle sizes provide more opportunity for workfow parallelism and load balancing. In this work, we introduce a domain-specifc data service that decouples that constraint so that data can be accessed and processed in its natural granularity while still maintaining I/O effciency. By removing the constraints introduced by fle handling we are able to obtain better scaling and make effcient use of more cores for processing a fxed-sized data sample. We demonstrate the improved scalability by using an application developed in the fle-based paradigm and comparing it to a version modifed to use HEPnOS.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Performance Portability of HPC Discovery Science Software: Fusion Energy Turbulence Simulations at Extreme Scale

As HPC R&D moves forward on a variety of “path to exascale” architectures today, an associated objective is to demonstrate performance portability of discovery-science-capable software. Important application domains, such as Magnetic Fusion Energy (MFE), have improved modelling of increasingly complex physical systems -- especially with respect to reducing “time-to-solution” as well as “energy to solution.” The emergence of new insights on confinement scaling in MFE systems has been aided significantly by efficient software capable of harnessing powerful supercomputers to carry out simulations with unprecedented resolution and temporal duration to address increasing problem sizes. Specifically, highly scalable particle-in-cell (PIC) programing methodology is used in this paper to demonstrate how modern scientific applications can achieve efficient architecture-dependent optimizations of performance scaling and code portability for path-to-exascale platforms.

97 MATHEMATICS AND COMPUTING↗

High Performance Computing Modeling Advances Accelerator Science for High-Energy Physics

The development and optimization of particle accelerators are essential for advancing our understanding of the properties of matter, energy, space, and time. Particle accelerators are complex devices whose behavior involves many physical effects on multiple scales. Therefore, advanced computational tools utilizing high-performance computing are essential for accurately modeling them. In the past decade, the US Department of Energy's SciDAC program has produced accelerator-modeling tools that have been employed to tackle some of the most difficult accelerator science problems. The authors discuss the Synergia framework and its applications to high-intensity particle accelerator physics. Synergia is an accelerator simulation package capable of handling the entire spectrum of beam dynamics simulations. Our authors present Synergia's design principles and its performance on HPC platforms.

43 PARTICLE ACCELERATORS↗

Massively Parallel Bayesian Model Calibration and Uncertainty Quantification with Applications to Nuclear Fuels and Materials

The U.S. Department of Energy (DOE)’s Nuclear Energy Advanced Modeling and Simulation (NEAMS) program aims to develop predictive capabilities by applying computational methods to the analysis and design of advanced reactor and fuel cycle systems. This program has been providing engineering-scale support for the development of BISON, a high-fidelity and high-resolution fuel performance tool. Fuel behavior in a nuclear reactor is governed by a complex network of mechanisms interacting with various other physics aspects in the reactor system. Any model developed to represent the fuel behavior will likely be idealized resulting in uncertainties in their predictions compared to the observed data. As such, this report was motivated by the need to identify the sources of uncertainties and quantify and propagate them through the fuel model outputs. Such quantification of uncertainties will establish a level of model trustworthiness, identify approaches to improve the model trustworthiness, and even guide optimal experiment design for maximal information gain. To accomplish the uncertainty quantification for computational models, this report has relied on the Bayesian framework which provides probabilistic treatment of models their inputs and outputs. The current state-of-the-art on performing Bayesian Uncertainty Quantification (UQ) for nuclear engineering models using High Performance Computing (HPC) resources have been reviewed. Implementation of capabilities for massively parallel Bayesian UQ in Multiphysics Object-Oriented Simulation Environment (MOOSE) is discussed. Several verification cases are discussed to verify the accuracy of the quantified uncertainties using the developed computational capabilities in MOOSE. Then, the problem of quantifying the uncertainties in TRI-Structural isOtropic (TRISO) fuel silver release is addressed. For the first time, the uncertainties arising from the TRISO Fission Gas Release (FGR) model due to model inadequacy and experimental noise are quantified. Also, the Bayesian capabilities are applied to the calibration of the MATPRO creep model, a widely used model in several fuel assessment cases. The impact of the prediction uncertainties in the MATPRO model on the fuel cladding behavior as part of the TRIBULATION assessment case (which is an integral effects case) is investigated. This report concludes with a discussion on the future work for the UQ for computational models.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Costa - Introduction to 2015 Annual Report

In parallel with Sandia National Laboratories having two major locations (NM and CA), along with a number of smaller facilities across the nation, so too is the distribution of scientific, engineering and computing resources. As a part of Sandia’s Institutional Computing Program, CA site-based Sandia computer scientists and engineers have been providing mission and research staff with local CA resident expertise on computing options while also focusing on two growing high performance computing research problems. The first is how to increase system resilience to failure, as machines grow larger, more complex and heterogeneous. The second is how to ensure that computer hardware and configurations are optimized for specialized data analytical mission needs within the overall Sandia computing environment, including the HPC subenvironment. All of these activities support the larger Sandia effort in accelerating development and integration of high performance computing into national security missions. Sandia continues to both promote national R&D objectives, including the recent Presidential Executive Order establishing the National Strategic Computing Initiative and work to ensure that the full range of computing services and capabilities are available for all mission responsibilities, from national security to energy to homeland defense.

97 MATHEMATICS AND COMPUTING↗

Scalable Deep-Learning-Accelerated Topology Optimization for Additively Manufactured Materials

Topology optimization (TO) is a popular and powerful computational approach for designing novel structures, materials, and devices. Two computational challenges have limited the applicability of TO to a variety of industrial applications. First, a TO problem often involves a large number of design variables to guarantee sufficient expressive power. Second, many TO problems require a large number of expensive physical model simulations, and those simulations cannot be parallelized. To address these issues, we propose a general scalable deep-learning (DL) based TO framework, referred to as SDL-TO, which utilizes parallel schemes in high performance computing (HPC) to accelerate the TO process for designing additively manufactured (AM) materials. Unlike the existing studies of DL for TO, our framework accelerates TO by learning the iterative history data and simultaneously training on the mapping between the given design and its gradient. The surrogate gradient is learned by utilizing parallel computing on multiple CPUs incorporated with a distributed DL training on multiple GPUs. The learned TO gradient enables a fast online update scheme instead of an expensive update based on the physical simulator or solver. Using a local sampling strategy, we achieve to reduce the intrinsic high dimensionality of the design space and improve the training accuracy and the scalability of the SDL-TO framework. The method is demonstrated by benchmark examples and AM materials design for heat conduction. The proposed SDL-TO framework shows competitive performance compared to the baseline methods but significantly reduces the computational cost by a speed up of around 8.6x over the standard TO implementation.

Bi, Sirui↗