Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “high performance computing systems”

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 1,153 records · Page 64

Improving High-Performance Sparse Libraries Using Compiler-Assisted Specialization: A PETSc Case Study

Computers with hardware accelerators, also referred to as hybrid-core systems, speedup applications by offloading certain compute operations that can run faster on accelerators. Thus, it is not surprising that many of top500 supercomputers use accelerators. However, in addition to procurement cost, significant programming and porting effort is required to realize the potential benefit of such accelerators. Hence, before building such a system it is prudent to answer the question 'what is the projected performance benefit from accelerators for the workloads of interest?'. We address this question by way of a performance-modeling framework that predicts realizable application performance on accelerators rapidly and accurately without going to the considerable effort of porting and tuning. The modeling framework first automatically identifies commonly found compute patterns in scientific applications which we term idioms, which may benefit by accelerator technology. Next the framework models the predicted speedup of those idioms if they were to be ported to and run on hardware accelerators. As a proof of concept we characterize two kinds of accelerators 1) the FPGA accelerators on a Convey HC-1 system and 2) an NVIDIA FERMI GPU accelerator. We model performance of the idioms gather/scatter and stream and our predictions show that where these occur in two full-scale HPC applications, Milc and HYCOM, gather/scatter speeds up by as much as 15X, and stream by as much as 14X, whereas the overall compute time of Milc improves by 3.4% and HYCOM by 20%.

Computer Science↗

A High Performance Computational Framework for Dynamic Security Assessment under Uncertainty

Dynamic security assessment (DSA) is a critical function to evaluate power grids’ capability to survive the transition caused by a set of disturbances to an acceptable steady-state condition. Its computational burden is heavy. With the challenges brought by renewable energy and new smart grid technologies, DSA under uncertainty has to be considered to study the impact of forecast errors on DSA simulation, which further increases the computational burden. To address this challenge, this paper presents a high performance computational framework to support DSA simulation user uncertainty. The computational framework aims to provide a seamless workflow that links data from high performance computing, statistical analysis, to visualization so that a problem can be easily expressed in a way compatible with different functions. It also enables software compatibility such that application development can be more efficient. Case study results of the ESCA60 system and a western U.S. size system show the advantages and efficiency of the computational framework.

dynamic security assessment, high performance comp↗

Phoebus: Network Middleware for Next-Generation Network Computing

The Phoebus project investigated algorithms, protocols, and middleware infrastructure to improve end-to-end performance in high speed, dynamic networks. The Phoebus system essentially serves as an adaptation point for networks with disparate capabilities or provisioning. This adaptation can take a variety of forms including acting as a provisioning agent across multiple signaling domains, providing transport protocol adaptation points, and mapping between distributed resource reservation paradigms and the optical network control plane. We have successfully developed the system and demonstrated benefits. The Phoebus system was deployed in Internet2 and in ESnet, as well as in GEANT2, RNP in Brazil and over international links to Korea and Japan. Phoebus is a system that implements a new protocol and associated forwarding infrastructure for improving throughput in high-speed dynamic networks. It was developed to serve the needs of large DOE applications on high-performance networks. The idea underlying the Phoebus model is to embed Phoebus Gateways (PGs) in the network as on-ramps to dynamic circuit networks. The gateways act as protocol translators that allow legacy applications to use dedicated paths with high performance.

99 GENERAL AND MISCELLANEOUS↗

AI-Science for Performance Optimization and Diagnosis of Science Instrument Federations

Next generation of science workflows are expected to be executed over complex federations composed of supercomputers, science instruments, storage systems and networks, with new additions of the edge and cloud systems and services. The sheer complexity of these multi-domain federations makes it hard to manage them and optimize their performance, as small impedance mismatches (that can dynamically develop between systems) could drastically degrade the entire federation performance. Recent proliferation of Software Defined Everything (SDX) technologies combined with containerization frameworks provide custom instruments that can monitor and collect critical measurements at various levels to support diagnoses and performance optimization; but their data too enormous for human operators and analysts to process and generate decisions. Machine Learning (ML) methods that extract critical parameters, relationships and trends from the data offer general solutions. Artificial Intelligence (AI) and ML methods must be custom-developed for these problems based on solid, rigorous foundations, since black-box approaches are often ineffective and unsound.We propose to develop comprehensive AI-Science for the performance of science federations to (i) monitor and control storage, networks, experiments, and computing systems across multiple domains via softwarization layers, at speeds and scales orders of magnitude superior to current practice, (ii) optimally realize and orchestrate complex workflows with high performance by using dynamic state and performance estimation methods, and (iii) aggregate measurements across sites and time to develop infrastructure-level profiles, optimizations and diagnoses using AI-Science based on foundational principles from ML, game theory, and information fusion areas.

Rao, Nageswara↗

IOrchestrator: improving the performance of multi-node I/O systems via inter-server coordination

A cluster of I/O nodes and a parallel file system are often used to provide high-throughput I/O service to a parallel compute cluster. To exploit I/O parallelism parallel file systems stripe file data across the I/O nodes. While this practice is effective in serving asynchronous requests, it may break individual program's spatial locality, which can seriously degrade I/O performance when the I/O nodes concurrently serve synchronous requests from multiple I/O-intensive programs. In this paper we propose a scheme, Orchestrator, to improve the I/O performance of multi-I/O-node systems by orchestrating I/O services among programs when such inter-I/O-node coordination is dynamically determined to be cost effective. We have implemented IOrchestrator in the PVFS2 parallel file system. Our experiments with representative parallel benchmarks show that IOrchestrator can significantly improve I/O performance - by up to a factor of 2.5 - delivered by a cluster of I/O nodes servicing concurrently-running parallel programs. Notably, we have not observed any scenarios in which the use of IOrchestrator causes significant performance degradation.

97 MATHEMATICS AND COMPUTING↗

MarFS, a Near-POSIX Interface to Cloud Objects

The engineering forces driving development of “cloud” storage have produced resilient, cost-effective storage systems that can scale to 100s of petabytes, with good parallel access and bandwidth. These features would make a good match for the vast storage needs of High-Performance Computing datacenters, but cloud storage gains some of its capability from its use of HTTP-style Representational State Transfer (REST) semantics, whereas most large datacenters have legacy applications that rely on POSIX file-system semantics. MarFS is an open-source project at Los Alamos National Laboratory that allows us to present cloud-style object-storage as a scalable near-POSIX file system. We have also developed a new storage architecture to improve bandwidth and scalability beyond what’s available in commodity object stores, while retaining their resilience and economy. Additionally, we present a scheme for scaling the POSIX interface to allow billions of files in a single directory and trillions of files in total.

97 MATHEMATICS AND COMPUTING↗

Optimizing Management of Persistent Data Structures in High-Performance Analytics

Large-scale data analytics workflows ingest massive input data into various data structures, including graphs and key-value datastores. These data structures undergo multiple transformations and computations and are typically reused in incremental and iterative analytics workflows. Persisting in-memory views of these data structures enables reusing them beyond the scope of a single program run while avoiding repetitive raw data ingestion overheads. Memory-mapped I/O enables persisting in-memory data structures without data serialization and deserialization overheads. However, memory-mapped I/O lacks the key feature of persisting consistent snapshots of these data structures for incremental ingestion and processing. The obstacles to efficient virtual memory snapshots using memory-mapped I/O include background writebacks outside the application’s control, and the significantly high storage footprint of such snapshots. To address these limitations, we present Privateer, a memory and storage management tool that enables storage-efficient virtual memory snapshotting while also optimizing snapshot I/O performance. Here, we integrated Privateer into Metall, a state-of-the-art persistent memory allocator for C++, and the Lightning Memory-Mapped Database (LMDB), a widely-used key-value datastore in data analytics and machine learning. Privateer optimized application performance by 1.22× when storing data structure snapshots to node-local storage, and up to 16.7× when storing snapshots to a parallel file system. Privateer also optimizes storage efficiency of incremental data structure snapshots by up to 11× using data deduplication and compression.

Computer science↗

The impact of low-Z and high-Z metal implants in IMRT: A Monte Carlo study of dose inaccuracies in commercial dose algorithms

Purpose: The aim of the study was to evaluate the dosimetric impact of low-Z and high-Z metallic implants on IMRT plans. Methods: Computed tomography (CT) scans of three patients were analyzed to study effects due to the presence of Titanium (low-Z), Platinum and Gold (high-Z) inserts. To eliminate artifacts in CT images, a sinogram-based metal artifact reduction algorithm was applied. IMRT dose calculations were performed on both the uncorrected and corrected images using a commercial planning system (convolution/superposition algorithm) and an in-house Monte Carlo platform. Dose differences between uncorrected and corrected datasets were computed and analyzed using gamma index (Pγ{sub <1}) and setting 2 mm and 2% as distance to agreement and dose difference criteria, respectively. Beam specific depth dose profiles across the metal were also examined. Results: Dose discrepancies between corrected and uncorrected datasets were not significant for low-Z material. High-Z materials caused under-dosage of 20%–25% in the region surrounding the metal and over dosage of 10%–15% downstream of the hardware. Gamma index test yielded Pγ{sub <1}>99% for all low-Z cases; while for high-Z cases it returned 91% < Pγ{sub <1}< 99%. Analysis of the depth dose curve of a single beam for low-Z cases revealed that, although the dose attenuation is altered inside the metal, it does not differ downstream of the insert. However, for high-Z metal implants the dose is increased up to 10%–12% around the insert. In addition, Monte Carlo method was more sensitive to the presence of metal inserts than superposition/convolution algorithm. Conclusions: The reduction in terms of dose of metal artifacts in CT images is relevant for high-Z implants. In this case, dose distribution should be calculated using Monte Carlo algorithms, given their superior accuracy in dose modeling in and around the metal. In addition, the knowledge of the composition of metal inserts improves the accuracy of the Monte Carlo dose calculation significantly.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

US Department of Energy, Office of Science High Performance Computing Facility Operational Assessment 2019 Oak Ridge Leadership Computing Facility

Oak Ridge National Laboratory's (ORNL's) Leadership Computing Facility (OLCF) continues to surpass its operational target goals: supporting users; delivering fast, reliable computational ecosystems; creating innovative solutions for high performance computing (HPC) needs; and managing risks, safety, and security associated with operating some of the most powerful computers in the world. The results can be seen in the cutting-edge science conducted by users and the praise from the research community. Calendar year (CY) 2019 was a big year as OLCF staff ran five world-class resources (the leadershipclass computers Titan and Summit, the large analysis cluster called Eos, and the massive parallel filesystems called Atlas and Alpine)) and also began power and cooling upgrades for a 2021 exascale system called Frontier. While continuing exceptional operation of Titan, Eos, and Rhea, the OLCF released the Summit supercomputer for production on January 1, 2019. Summit debuted as the most capable and efficient system in its class and has been recognized as the most powerful system in the world for its performance on both the high performance linpack (HPL) and conjugate gradient (HPCG) benchmark applications since June 2018 according to TOP500. Summit represents the culmination of a multiyear effort between the OLCF, IBM, NVIDIA, and Mellanox to deliver a system that is unmatched for modeling, simulation, data analysis, and learning. To hit the ground running with science-ready applications on day one, application teams worked closely with the OLCF through the Center for Accelerated Application Readiness (CAAR) program for years in advance of the Summit deployment. CY 2019 was filled with outstanding results and accomplishments: a very high rating from users on overall satisfaction for the sixth year in a row; a tremendous amount of core-hours delivered to researchers from two leadership-class systems; and success in delivering on the allocation split of roughly 60%, 30%, and 10% of core-hours offered for the Innovative and Novel Computational Impact on Theory and Experiment (INCITE), Advanced Scientific Computing Research Leadership Computing Challenge (ALCC), and Director's Discretionary (DD) programs, respectively (see Operational Performance section). These accomplishments, coupled with the high utilization rates (overall and capability usage), represent the fulfillment of the promise of both leadership-class machines: efficient facilitation of leadership-class computational applications. Table ES.1 presents a summary of the 2019 OLCF metric targets and the associated results. More information can be found in the Operational Performance section for each OLCF resource. The scientific accomplishments of OLCF users are a strong indication of long-term operational success, with publications this year in such notable journals and publications as Nature, Nature Physics, Nature Plants, Physical Review X, Journal of the American Physical Society, Cell, Nano Letters, and Trends in Biotechnology. Crucial domain-specific discoveries facilitated by resources at the OLCF are described in the High Performance Computing Facility Operational Assessment 2019 Oak Ridge Leadership Computing Facility (OAR) Strategic Results section. For example, researchers used Summit to pinpoint and understand the production of proteins from genetic information, including mutations and the functional expression of disease (Section 8.2).

97 MATHEMATICS AND COMPUTING↗

Computational framework for modeling of physical process

Techniques, systems, and devices are described for providing a computational frame for estimating high-dimensional stochastic behaviors. In one exemplary aspect, a method for performing numerical estimation includes receiving a set of measurements of a stochastic behavior. The set of correlated measurements follows a non-standard probability distribution and is non-linearly correlated. Also, a non-linear relationship exists between a set of system variables that describes the stochastic behavior and a corresponding set of measurements. The method includes determining, based on the set of measurements, a numerical model of the stochastic behavior. The numerical model comprises a feature space comprising non-correlated features corresponding to the stochastic behavior. The non-correlated features have a dimensionality of M and the set of measurements has a dimensionality of N, M being smaller than N. The method includes generating a set of approximated system variables corresponding to the set of measurements based on the numerical model.

Chen, Xiao↗

Computational framework for modeling of physical process

Techniques, systems, and devices are described for providing a computational frame for estimating high-dimensional stochastic behaviors. In one exemplary aspect, a method for performing numerical estimation includes receiving a set of measurements of a stochastic behavior. The set of correlated measurements follows a non-standard probability distribution and is non-linearly correlated. Also, a non-linear relationship exists between a set of system variables that describes the stochastic behavior and a corresponding set of measurements. The method includes determining, based on the set of measurements, a numerical model of the stochastic behavior. The numerical model comprises a feature space comprising non-correlated features corresponding to the stochastic behavior. The non-correlated features have a dimensionality of M and the set of measurements has a dimensionality of N, M being smaller than N. The method includes generating a set of approximated system variables corresponding to the set of measurements based on the numerical model.

Chen, Xiao↗

Operating System Support for Fine-Grained Task Migration

With the end of clock scaling and the limited power budget available (20-30MW), future supercomputers will meet exascale performance primarily through a higher level of parallelism. Current operating (OS) and runtime (RT) systems are designed for the classical SMP model and based on the static and coarse-grained process/thread paradigm. They do not provide the required level of flexibility, especially within a single compute node, to meet the requirements imposed by exascale systems in terms of power/energy efficiency, resilience, managing concurrency and performance portability. The high level of concurrency poses new challenges specific to exascale systems that need to be addressed by novel solutions. In particular, requiring the user to manage billions of concurrent threads could easily result in poor data locality, clogged interconnection networks, unchecked propagation of soft errors, and lack of control over power/energy consumption. Equally important, managing such level of concurrency interferes with the user’s focus on the application and the algorithm. We envision that the computation will be encapsulated into fine-grained tasks that can be isolated and protected from the other tasks running in the system. Whenever a task needs to work on some data that is not stored on the local node, the OS/RT allows the task to be migrated to the node that owns the data. Each task is associated with a contained state (set or architectural registers, stack frame, running node, etc.) that describes the progress of the task and that should be moved together with the task’s code. We believe system support for task migration is a fundamental function that can ease the job of tackling several of the exascale challenges.

Gioiosa, Roberto↗

Large File System Backup: NERSC Global File System Experience

NERSC's Global File system (NGF), accessible from all compute systems at NERSC, holds files and data from many scientific projects. A full backup of this file system to our High Performance Storage System (HPSS) is performed periodically. Disk storage usage by projects at NERSC has grown seven fold over a two year period, from ~;;20TB in June 2006 to ~;;140 TB in June 2008. The latest full backup took about 13 days and more than 200 T10k tape cartridges (.5 TB capacity). Petabyte file systems are becoming a reality in the next few years and the existing utilities are already strained in handling backup tasks.

97 MATHEMATICS AND COMPUTING↗

Multiscale modeling and characterization for performance and safety of lithium-ion batteries

Lithium-ion batteries are highly complex electrochemical systems whose performance and safety are governed by coupled nonlinear electrochemical-electrical-thermal-mechanical processes over a range of spatiotemporal scales. Gaining an understanding of the role of these processes as well as development of predictive capabilities for design of better performing batteries requires synergy between theory, modeling, and simulation, and fundamental experimental work to support the models. This paper presents the overview of the work performed by the authors aligned with both experimental and computational efforts. In this paper, we describe a new, open source computational environment for battery simulations with an initial focus on lithium-ion systems but designed to support a variety of model types and formulations. This system has been used to create a three-dimensional cell and battery pack models that explicitly simulate all the battery components (current collectors, electrodes, and separator). The models are used to predict battery performance under normal operations and to study thermal and mechanical safety aspects under adverse conditions. This paper also provides an overview of the experimental techniques to obtain crucial validation data to benchmark the simulations at various scales for performance as well as abuse. We detail some initial validation using characterization experiments such as infrared and neutron imaging and micro-Raman mapping. In addition, we identify opportunities for future integration of theory, modeling, and experiments.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Linear and Nonlinear Solvers for Simulating Multiphase Flow within Large-Scale Engineered Subsurface Systems

Simulation of multiphase flow in the subsurface is well-known to be computationally challenging. While there have been many studies that have explored approaches to overcoming these challenges, they often utilize relatively simple case studies. In this paper, we focus on the unique numerical challenges posed by modeling large-scale engineered subsurface systems, characterized by discrete features embedded in a heterogeneous natural subsurface setting. The man-made features such as shafts, tunnels, and barriers often cause multiple challenges in modeling the domain for multiphase porous media flow. This flow scenario can have a wide range of applications such as nuclear waste repositories, enhanced recovery of a petroleum reservoir, geothermal engineering, and carbon sequestration. An example of these severe numerical challenges is the case of performance assessment (PA) for Waste Isolation Pilot Plant (WIPP), the only operating deep geological repository in the US, which simulates extreme material properties of bedded salt rock formation and extreme contrast due to open excavation next to the formation. The models have extremes not only of permeability and porosity but also of the constitutive models needed for multiphase flow; additionally, they have process models like salt creep closure reducing porosity over time, fracturing in clay and anhydrite interbeds of the bedded salt, gas generation from the waste materials, and unintentional human borehole intrusions in some scenarios. Numerical simulations require the solution of coupled systems of nonlinear PDEs; in our work, we use the open-source simulator PFLOTRAN which is based on Finite Volume discretization. The solution of the nonlinear equations requires use of the Newton-Raphson iteration at each time step, which entails the solution of the linearized Jacobian system at each iteration. The effects of all the processes (i.e., large number of unknowns, highly nonlinear constitutive relations, large contrasts in material properties in short distances) lead to an ill-conditioned Jacobian matrix that severely challenges traditional linear solver, i.e., stabilized biconjugate gradient with block Jacobi incomplete LU preconditioner (BCGS-ILU) leading to non-convergence for traditional Newton-Raphson nonlinear solver causing unacceptably long computation time for each model. This paper presents linear solvers such as constrained pressure residual (CPR) two-stage preconditioner with alternate-block-factorization (ABF) and quasi- implicit pressure and explicit saturation (QIMPES) decouplers and flexible generalized residual solver (FGMRES). The new general-purpose nonlinear solver, Newton trust-region dogleg Cauchy (NTRDC), is also introduced to resolve extreme nonlinearities in the models. We demonstrate the effectiveness of each method relative to the default BCGS-Newton solver. The two best cases had nearly 50 times speed-up and achieved completion of a simulation in 14 hours that never completed due to non-convergence with the default solver. We also investigate the strong scalability of each method and discuss some of the deficiencies found for Block Jacobi preconditioner using parallel domain decomposition, and node packing effects of modern processor architecture.

Preconditioner, Nonlinear, Porous media, Multiphas↗

Evidence of Skewness and Sub-Gaussian Character in Temperature-Dependent Distributions of One Million Electronic Excitation Energies in PbS Quantum Dots

Obtaining statistical distributions by sampling a large number of conformations is vital for an accurate description of temperature-dependent properties of chemical systems. However, constructing distributions with 105??106 samples is computationally challenging because of the prohibitively high computational cost of performing rst-principles quantum-mechanical calculations. In this work, we present a new technique called the Eective Stochastic Potential Conguration Interaction Singles (ESP-CIS) method to obtain excitation energies. The ESP-CIS method uses random matrix theory for the construction of an eective stochastic representation of the Fock operator and combines it with linear-response theory. Here, we perform excitation calculations on PbS quantum dots (0.75-1.75 nm) at temperatures of 200-400 K. Results from a total of 15 million excitation energy calculations (1 million samples for each of the three dots at ve dierent temperatures) revealed the distributions to be sub-Gaussian in nature with negative skewness. The distributions progressively became red-shifted with increasing temperature. The study shows that random distortions in structure from the minimum-energy geometry of these quantum dots are more likely to lower the excitation energies.

Scher, Jeremy A.↗

5G Energy FRAME: The Design and Implementation of Data, Model, and Use Case (Year 2 Report)

This report summarizes the Year 2 work of Pacific Northwest National Laboratory’s (PNNL’s) 5G Fabricated Resource and Asset Management Encompassment for energy infrastructure (Energy FRAME) project funded by the Department of Energy Office of Science’s Advanced Scientific Computing Research (ASCR) Program. In this report, newest 5G equipment testing results are presented, along with two 5G-enabled AI/ML examples for grid applications; in addition, the work flow of grid edge, cloud, and High Performance Computing (HPC) platform is introduced, to support and interface the cross-domain simulation for power system transmission, distribution, and communication networks. Last but not least, the outlook for Year 3 work and the overarching impact of 5G Energy FRAME work to a multitude of stakeholders are provided. Additional 5G performance data now is shared through the publicly available weblink, https://www.pnnl.gov/projects/5g-energy-frame/publications

24 POWER TRANSMISSION AND DISTRIBUTION↗