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 1,657 records · Page 92

An Optimizing Compiler for Petascale I/O on Leadership-Class Architectures

In high-performance computing (HPC), parallel I/O architectures usually have very complex hierarchies with multiple layers that collectively constitute an I/O stack, including high-level I/O libraries such as PnetCDF and HDF5, I/O middleware such as MPI-IO, and parallel file systems such as PVFS and Lustre. Our DOE project explored automated instrumentation and compiler support for I/O intensive applications. Our project made significant progress towards understanding the complex I/O hierarchies of high-performance storage systems (including storage caches, HDDs, and SSDs), and designing and implementing state-of-the-art compiler/runtime system technology that targets I/O intensive HPC applications that target leadership class machine. This final report summarizes the major achievements of the project and also points out promising future directions Two new sections in this report compared to the previous report are IOGenie and SSD/NVM-specific optimizations.

97 MATHEMATICS AND COMPUTING↗

Insights From the Software Design of a Multiphysics Multicomponent Scientific Code

Using simulations for scientific discovery requires that the software used in the simulations undergoes a rigorous design and development process similar to that of the lab instruments in the experimental sciences. To devise a good design methodology, it is critical to understand the requirements, constraints, and challenges. Furthermore, this article describes insights from the long-term stewardship of a multiphysics multicomponent software, FLASH, that was designed more than 20 years ago for astrophysics, now serves multiple communities, and has been successful in adapting to the changing world of high-performance computing.

97 MATHEMATICS AND COMPUTING↗

Timely Result-Data Offloading for Improved HPC Center Scratch Provisioning and Serviceability

Modern High-Performance Computing (HPC) centers are facing a data deluge from emerging scientific applications. Supporting large data entails a significant commitment of the highthroughput center storage system, scratch space. However, the scratch space is typically managed using simple purge policies, without sophisticated end-user data services to balance resource consumption and user serviceability. End-user data services such as offloading are performed using point-to-point transfers that are unable to reconcile center s purge and users delivery deadlines, unable to adapt to changing dynamics in the end-toend data path and are not fault-tolerant. Such inefficiencies can be prohibitive to sustaining high performance. In this paper, we address the above issues by designing a framework for the timely, decentralized offload of application result data. Our framework uses an overlay of user-specified intermediate and landmark sites to orchestrate a decentralized fault-tolerant delivery. We have implemented our techniques within a production job scheduler (PBS) and data transfer tool (BitTorrent). Our evaluation using both a real implementation and supercomputer job log-driven simulations show that: the offloading times can be significantly reduced (90.4% for a 5 GB data transfer); the exposure window can be minimized while also meeting center-user Service Level Agreements.

99 GENERAL AND MISCELLANEOUS↗

DenKv: Addressing Design Trade-offs of Key-value Stores for Scientific Applications

High-performance computing (HPC) facilities have employed flash-based storage tier near to compute nodes to absorb high I/O demand by HPC applications during periodic system-level checkpoints. To accelerate these checkpoints, proxy-based distributed key-value stores (PD-KVS) gained particular attention for their flexibility to support multiple backends and different network configurations. PD-KVS rely internally on monolithic KVS, such as LevelDB or RocksDB, to exploit the KV interface and query support. However, PD-KVS are unaware of the high redundancy factor in checkpoint data, which can be up to GBs to TBs, and therefore, tend to generate high write and space amplification on these storage layers. In this paper, we propose DenKv which is deduplication-extended node-local LSM-tree-based KVS. DenKv employs asynchronous partially inline dedup (APID) and aims to maintain the performance characteristics of LSM-tree-based KVS while reducing the write and space amplification problems. We implemented DenKv atop BlobDB and showed that our proposed solution maintains performance while reducing write amplification up to 2× and space amplification by 4× on average.

Khan, Awais↗

Turbomachine Sealing and Secondary Flows - Part 3: Review of Power-Stream Support, Unsteady Flow Systems, Seal and Disk Cavity Flows, Engine Externals, and Life and Reliability Issues - Part 3

The issues and components supporting the engine power stream are reviewed. It is essential that companies pay close attention to engine sealing issues, particularly on the high-pressure spool or high-pressure pumps. Small changes in these systems are reflected throughout the entire engine. Although cavity, platform, and tip sealing are complex and have a significant effect on component and engine performance, computational tools (e.g., NASA-developed INDSEAL, SCISEAL, and ADPAC) are available to help guide the designer and the experimenter. Gas turbine engine and rocket engine externals must all function efficiently with a high degree of reliability in order for the engine to run but often receive little attention until they malfunction. Within the open literature statistically significant data for critical engine components are virtually nonexistent; the classic approach is deterministic. Studies show that variations with loading can have a significant effect on component performance and life. Without validation data they are just studies. These variations and deficits in statistical databases require immediate attention.

Hendricks, R. C.↗

Advanced co-simulation for computer-aided process design and optimization of fossil energy systems with carbon capture

In this paper, we describe recent progress toward developing an Advanced Process Engineering Co-Simulator (APECS) for use in computer-aided design and optimization of fossil energy systems with carbon capture. The APECS system combines process simulation with multiphysicsbased equipment simulations, such as those based on computational fluid dynamics. These co-simulation capabilities enable design engineers to optimize overall process performance with respect to complex thermal and fluid flow phenomena arising in key plant equipment items. This paper also highlights ongoing co-simulation R&D activities in areas such as reduced order modeling, knowledge management, stochastic analysis and optimization, and virtual plant co-simulation. Continued progress in co-simulation technology— through improved integration, solution, deployment, and analysis —will have profound positive impacts on the design and optimization of high-efficiency, near-zero emission fossil energy systems.

97 MATHEMATICS AND COMPUTING↗

Designing a Scalable Fault Tolerance Model for High Performance Computational Chemistry: A Case Study with Coupled Cluster Perturbative Triples

In the last couple of decades, the massive computational power provided by the most modern supercomputers has resulted in simulation of higher order computational chem- istry methods, previously considered intractable. As the system sizes continue to increase, computational chemistry domain continues to escalate this trend using parallel comput- ing with programming models such as Message Passing Interface (MPI) and Partitioned Global Address Space (PGAS) programming models such as Global Arrays. The ever increasing scale of these supercomputers comes at a cost of reduced mean time between failures, currently in the order of days, and projected to be in the order of hours for up- coming extreme scale systems. While traditional disk based checkpointing methods are ubiquitous for storing intermediate solutions, they suffer from high overhead of writing and recovering from checkpoints. In practice, checkpointing itself often brings the system down. Clearly, methods beyond checkpointing are imperative to handling the aggravating issue of reducing MTBF. In this paper, we address this challenge by designing and im- plementing an efficient fault tolerant version of coupled cluster method with NWChem, using in memory data redundancy. We present the challenges associated with our de- sign including efficient data storage model, maintenance of at least one consistent data copy and the recovery process. Our performance evaluation without faults shows that the current design exhibits negligible overhead. In the presence of a fault, the proposed design incurs negligible overhead in comparison to the state of the art implementation without faults.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

High-Performance Algorithm for Solving the Diagnosis Problem

An improved method of model-based diagnosis of a complex engineering system is embodied in an algorithm that involves considerably less computation than do prior such algorithms. This method and algorithm are based largely on developments reported in several NASA Tech Briefs articles: The Complexity of the Diagnosis Problem (NPO-30315), Vol. 26, No. 4 (April 2002), page 20; Fast Algorithms for Model-Based Diagnosis (NPO-30582), Vol. 29, No. 3 (March 2005), page 69; Two Methods of Efficient Solution of the Hitting-Set Problem (NPO-30584), Vol. 29, No. 3 (March 2005), page 73; and Efficient Model-Based Diagnosis Engine (NPO-40544), on the following page. Some background information from the cited articles is prerequisite to a meaningful summary of the innovative aspects of the present method and algorithm. In model-based diagnosis, the function of each component and the relationships among all the components of the engineering system to be diagnosed are represented as a logical system denoted the system description (SD). Hence, the expected normal behavior of the engineering system is the set of logical consequences of the SD. Faulty components lead to inconsistencies between the observed behaviors of the system and the SD. Diagnosis the task of finding faulty components is reduced to finding those components, the abnormalities of which could explain all the inconsistencies. The solution of the diagnosis problem should be a minimal diagnosis, which is a minimal set of faulty components. The calculation of a minimal diagnosis is inherently a hard problem, the solution of which requires amounts of computation time and memory that increase exponentially with the number of components of the engineering system. Among the developments to reduce the computational burden, as reported in the cited articles, is the mapping of the diagnosis problem onto the integer-programming (IP) problem. This mapping makes it possible to utilize a variety of algorithms developed previously for IP to solve the diagnosis problem. In the IP approach, the diagnosis problem can be formulated as a linear integer optimization problem, which can be solved by use of well-developed integer-programming algorithms. This concludes the background information.

Fijany, Amir↗

HTMT-class Latency Tolerant Parallel Architecture for Petaflops Scale Computation

Computational Aero Sciences and other numeric intensive computation disciplines demand computing throughputs substantially greater than the Teraflops scale systems only now becoming available. The related fields of fluids, structures, thermal, combustion, and dynamic controls are among the interdisciplinary areas that in combination with sufficient resolution and advanced adaptive techniques may force performance requirements towards Petaflops. This will be especially true for compute intensive models such as Navier-Stokes are or when such system models are only part of a larger design optimization computation involving many design points. Yet recent experience with conventional MPP configurations comprising commodity processing and memory components has shown that larger scale frequently results in higher programming difficulty and lower system efficiency. While important advances in system software and algorithms techniques have had some impact on efficiency and programmability for certain classes of problems, in general it is unlikely that software alone will resolve the challenges to higher scalability. As in the past, future generations of high-end computers may require a combination of hardware architecture and system software advances to enable efficient operation at a Petaflops level. The NASA led HTMT project has engaged the talents of a broad interdisciplinary team to develop a new strategy in high-end system architecture to deliver petaflops scale computing in the 2004/5 timeframe. The Hybrid-Technology, MultiThreaded parallel computer architecture incorporates several advanced technologies in combination with an innovative dynamic adaptive scheduling mechanism to provide unprecedented performance and efficiency within practical constraints of cost, complexity, and power consumption. The emerging superconductor Rapid Single Flux Quantum electronics can operate at 100 GHz (the record is 770 GHz) and one percent of the power required by convention semiconductor logic. Wave Division Multiplexing optical communications can approach a peak per fiber bandwidth of 1 Tbps and the new Data Vortex network topology employing this technology can connect tens of thousands of ports providing a bi-section bandwidth on the order of a Petabyte per second with latencies well below 100 nanoseconds, even under heavy loads. Processor-in-Memory (PIM) technology combines logic and memory on the same chip exposing the internal bandwidth of the memory row buffers at low latency. And holographic storage photorefractive storage technologies provide high-density memory with access a thousand times faster than conventional disk technologies. Together these technologies enable a new class of shared memory system architecture with a peak performance in the range of a Petaflops but size and power requirements comparable to today's largest Teraflops scale systems. To achieve high-sustained performance, HTMT combines an advanced multithreading processor architecture with a memory-driven coarse-grained latency management strategy called "percolation", yielding high efficiency while reducing the much of the parallel programming burden. This paper will present the basic system architecture characteristics made possible through this series of advanced technologies and then give a detailed description of the new percolation approach to runtime latency management.

Sterling, Thomas↗

Development of experimental and computational frameworks to predict subcooled flow boiling in the LANL Isotope Production Facility

Cooling is crucial to maintain the integrity of target systems in isotope production facilities. At Los Alamos National Laboratory (LANL)’s Isotope Production Facility (IPF), multiple encapsulated targets are stacked and irradiated in tandem with a 100 MeV, ~250μA proton beam. To facilitate effective heat removal, these stacked targets are separated and cooled via a series of water channels. At these beam currents, this high-energy proton beam heats the target system, likely initiating subcooled flow boiling in the cooling channels. However, in-beam monitoring of the IPF target system is not possible due to the extreme radiation environment, and the necessarily significant shielding. To better understand high-power target performance, we developed ex-situ experimental and computational frameworks to predict the behavior of subcooled flow boiling at IPF. Subcooled flow boiling experiments on Inconel 625 samples under IPF conditions (2 bar pressure, 10 GPM flow rate (i.e., 2249 kg/m 2 /s), 85 K subcooling) revealed that IPF's average operating power is at the early stage of boiling with a heat transfer coefficient of 48,000 W/m 2 /s. The proposed modeling framework enables us to predict a complete boiling curve, i.e., single-phase heat transfer, onset of nucleate boiling, two-phase heat transfer, and critical heat flux (CHF), with specification of input boiling parameters up to intermediate heat flux levels. The estimated CHF under IPF conditions is 5.2 MW/m 2 . Experimental data under reduced conditions (2 bar pressure, 1.5 GPM flow rate (i.e., 337 kg/m 2 /s), 45 K subcooling) served as validation cases for the computational modeling. This computational model can be further extended to more complicated systems replicating the real IPF configuration, for instance, to study void distribution as a function of the incident proton beam profile and coolant velocity profile of multiple cooling channels. Finally, the proposed experimental and computational frameworks provide a means to better understand cooling systems in the isotope production facilities at different accelerators, where in-beam monitoring of the cooling process is not available.

07 ISOTOPE AND RADIATION SOURCES↗

Computations of Torque-Balanced Coaxial Rotor Flows

Interactional aerodynamics has been studied for counter-rotating coaxial rotors in hover. The effects of torque balancing on the performance of coaxial-rotor systems have been investigated. The three-dimensional unsteady Navier-Stokes equations are solved on overset grids using high-order accurate schemes, dual-time stepping, and a hybrid turbulence model. Computational results for an experimental model are compared to available data. The results for a coaxial quadcopter vehicle with and without torque balancing are discussed. Understanding interactions in coaxial-rotor flows would help improve the design of next-generation autonomous drones.

Torque↗

A Parallel Graph Environment for Real-World Data Analytics Workflows

Economic competitiveness and national security depend increasingly on the insightful analysis of large data sets. The diversity of real-world data sources and analytic workflows impose challenging hardware and software requirements for parallel graph platforms. The irregular nature of graph methods is not supported well by the deep memory hierarchies of conventional distributed systems, requiring new processor and runtime system designs to tolerate memory and synchronization latencies. Moreover, the efficiency of relational table operations and matrix computations are not attainable when data is stored in common graph data structures. In this paper, we present HAGGLE, a high-performance, scalable data analytics platform. The platform’s hybrid data model supports a variety of distributed, thread-safe data structures, parallel programming constructs, and persistent and streaming data. An abstract runtime layer enables us to map the stack to conventional, distributed computer systems with accelerators. The runtime uses multithreading, active messages, and data aggregation to hide memory and synchronization latencies on large-scale systems.

Graph Analytics↗

A massively parallel semi-Lagrangian algorithm for solving the transport equation

The scalar transport equation underpins many models employed in science, engineering, technology and business. Application areas include, but are not restricted to, pollution transport, weather forecasting, video analysis and encoding (the optical flow equation), options and stock pricing (the Black-Scholes equation) and spatially explicit ecological models. Unfortunately finding numerical solutions to this equation which are fast and accurate is not trivial. Moreover, finding such numerical algorithms that can be implemented on high performance computer architectures efficiently is challenging. In this paper the authors describe a massively parallel algorithm for solving the advection portion of the transport equation. We present an approach here which is different to that used in most transport models and which we have tried and tested for various scenarios. The approach employs an intelligent domain decomposition based on the vector field of the system equations and thus automatically partitions the computational domain into algorithmically autonomous regions. The solution of a classic pure advection transport problem is shown to be conservative, monotonic and highly accurate at large time steps. Additionally we demonstrate that the algorithm is highly efficient for high performance computer architectures and thus offers a route towards massively parallel application.

54 ENVIRONMENTAL SCIENCES↗

Machine learning the metastable phase diagram of covalently bonded carbon

Abstract Conventional phase diagram generation involves experimentation to provide an initial estimate of the set of thermodynamically accessible phases and their boundaries, followed by use of phenomenological models to interpolate between the available experimental data points and extrapolate to experimentally inaccessible regions. Such an approach, combined with high throughput first-principles calculations and data-mining techniques, has led to exhaustive thermodynamic databases (e.g. compatible with the CALPHAD method), albeit focused on the reduced set of phases observed at distinct thermodynamic equilibria. In contrast, materials during their synthesis, operation, or processing, may not reach their thermodynamic equilibrium state but, instead, remain trapped in a local (metastable) free energy minimum, which may exhibit desirable properties. Here, we introduce an automated workflow that integrates first-principles physics and atomistic simulations with machine learning (ML), and high-performance computing to allow rapid exploration of the metastable phases to construct “metastable” phase diagrams for materials far-from-equilibrium. Using carbon as a prototypical system, we demonstrate automated metastable phase diagram construction to map hundreds of metastable states ranging from near equilibrium to far-from-equilibrium (400 meV/atom). We incorporate the free energy calculations into a neural-network-based learning of the equations of state that allows for efficient construction of metastable phase diagrams. We use the metastable phase diagram and identify domains of relative stability and synthesizability of metastable materials. High temperature high pressure experiments using a diamond anvil cell on graphite sample coupled with high-resolution transmission electron microscopy (HRTEM) confirm our metastable phase predictions. In particular, we identify the previously ambiguous structure of n -diamond as a cubic-analog of diaphite-like lonsdaelite phase.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Multi-Fidelity Stochastic Economic Dispatch for Operating Low Carbon Power Grids

Grid operators can address the inherently stochastic nature of renewables by solving a two-stage stochastic programming model that minimizes the cost of dispatch decisions while accounting for the complex grid dynamics. It is common to use a sample average approximation to estimate the expectation of the second stage costs in this model. However, the large sample count needed for numerical accuracy makes effective modeling large-scale electric grids computationally intractable. We introduce a control variate multi-fidelity estimator for the second-stage recourse that enables high quality dispatch decisions in real-time with a reduced computational burden. We obtain a hierarchy of model fidelities by linearizing the AC power flow system representation to DC power flow, and by relaxing transmission and voltage network constraints. We evaluate the performance of our proposed method on a synthetic grid with 73 buses against a deterministic baseline with persistence forecast and a high-fidelity reference. Our analysis shows a computational speed-up of 7.62x with a minimal loss in accuracy. The multi-fidelity method is well suited to fidelity combinations that use a simplified network topology in their lower fidelity model and is an attractive option for applications where accurate grid modeling needed on a limited computational budget.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Quantum simulations of excited states with active-space downfolded Hamiltonians

Many-body techniques based on the double unitary coupled cluster ansatz (DUCC) can be used to downfold electronic Hamiltonians into low-dimensional active spaces. It can be shown that the resulting dimensionality reduced Hamiltonians are amenable for quantum computing. Recent studies performed for several benchmark systems using quantum phase estimation (QPE) algorithms demonstrated that these formulations can recover a significant portion of ground-state dynamical correlation effects that stem from the electron excitations outside of the active space. These results have also been confirmed in studies of ground-state potential energy surfaces using quantum simulators. In this letter, we study the effectiveness of the DUCC formalism in describing excited states. We also emphasize the role of the QPE formalism and its stochastic nature in discovering/identifying excited states or excited-state processes in situations when the knowledge about the true configurational structure of a sought after excited state is limited or postulated (due to the specific physics driving excited-state processes of interest). In this context, we can view the QPE algorithm as an engine for verifying various hypotheses for excited-state processes and providing statistically meaningful results that correspond to the electronic state(s) with the largest overlap with a postulated configurational structure. We illustrate these ideas on examples of strongly cor- related molecular systems, characterized by small energy gaps and high density of quasi-degenerate states around the Fermi level.

quantum chemistry, quantum computing, computationa↗

Sensitivity analysis of generic deep geologic repository with focus on spatial heterogeneity induced by stochastic fracture network generation

Geologic Disposal Safety Assessment Framework is a state-of-the-art simulation software toolkit for probabilistic post-closure performance assessment of systems for deep geologic disposal of nuclear waste developed by the United States Department of Energy. This paper presents a generic reference case and shows how it is being used to develop and demonstrate performance assessment methods within the Geologic Disposal Safety Assessment Framework that mitigate some of the challenges posed by high uncertainty and limited computational resources. Variance-based global sensitivity analysis is applied to assess the effects of spatial heterogeneity using graph-based summary measures for scalar and time-varying quantities of interest. Behavior of the system with respect to spatial heterogeneity is further investigated using ratios of water fluxes. This analysis shows that spatial heterogeneity is a dominant uncertainty in predictions of repository performance which can be identified in global sensitivity analysis using proxy variables derived from graph descriptions of discrete fracture networks. New quantities of interest defined using water fluxes proved useful for better understanding overall system behavior.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Energy scaling advantages of resistive memory crossbar based computation and its application to sparse coding

In this study, the exponential increase in data over the last decade presents a significant challenge to analytics efforts that seek to process and interpret such data for various applications. Neural-inspired computing approaches are being developed in order to leverage the computational properties of the analog, low-power data processing observed in biological systems. Analog resistive memory crossbars can perform a parallel read or a vector-matrix multiplication as well as a parallel write or a rank-1 update with high computational efficiency. For an N × N crossbar, these two kernels can be O(N) more energy efficient than a conventional digital memory-based architecture. If the read operation is noise limited, the energy to read a column can be independent of the crossbar size (O(1)). These two kernels form the basis of many neuromorphic algorithms such as image, text, and speech recognition. For instance, these kernels can be applied to a neural sparse coding algorithm to give an O(N) reduction in energy for the entire algorithm when run with finite precision. Sparse coding is a rich problem with a host of applications including computer vision, object tracking, and more generally unsupervised learning.

97 MATHEMATICS AND COMPUTING↗