Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data locality”

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 91 records · Page 5

Asynchronous in situ connected-components analysis for complex fluid flows

Fluid flow in porous media is at the heart of problems such as groundwater contamination and carbon sequestration, and presents an important challenge for scientific computing. For this class of problem, large three-dimensional simulations are performed to advance scientific inquiry. On massively parallel computing systems, the volume of data generated by such approaches can become a productivity bottleneck if the raw data generated from the simulation is analyzed in a post-processing step. We present a physics-based framework for in situ data reduction that is theoretically grounded in multiscale averaging theory. We show how task parallelism can be exploited to concurrently perform a variety of analysis tasks with data-dependent costs, including the generation of iso-surfaces, morphological analyses, and connected components analysis. A task management framework is constructed to launch asynchronous analysis threads, manage dependencies between different tasks, promote data locality and hide the impact of data transfers. The framework is applied to analyze GPU-based simulations of two-fluid flow in porous media, generating a set of averaged measures that represents the overall system behavior. We demonstrate how the approach can be applied to perform physically-consistent averaging over fluid subregions using connected components analysis. Simulations performed on Oak Ridge National Lab s Titan supercomputer are profiled to demonstrate the performance of the associated multi-threaded in situ analysis for typical production simulation of two-fluid flow.

Mcclure, James↗

Asynchronous In Situ Connected-Components Analysis for Complex Fluid flows, In: ISAV '16 Proceedings of the 2nd Workshop on In Situ Infrastructures for Enabling Extreme-scale Analysis and Visualization

The simulation of multiscale physics is an important challenge for scientific computing. For this class of problem, large three-dimensional simulations are performed to advance scientific inquiry. On massively parallel computing systems, the volume of data generated by such approaches can become a productivity bottleneck if the raw data generated from the simulation is analyzed in a post-processing step. To address this, we present a physics-based framework for in situ data reduction that is theoretically grounded in multiscale averaging theory. We show how task parallelism can be exploited to concurrently perform a variety of analysis tasks with data-dependent costs, including the generation of iso-surfaces, morphological analyses, and connected components analysis. All analyses are performed in parallel using distributed memory and use the same domain decomposition as the simulation. A task management framework is constructed to leverage available parallelism within a node for analysis. The capabilities of the framework are to launch asynchronous analysis threads, manage dependencies between different tasks, promote data locality and minimize the impact of data transfers. The framework is applied to analyze GPU-based simulations of two-fluid-phase flow in porous media, generating a set of averaged measures that represents the overall system behavior. We demonstrate how the approach can be applied to perform physically-consistent analysis over fluid sub-regions determined from connected components analysis. Simulations performed on Oak Ridge National Lab's Titan supercomputer are profiled to demonstrate the performance of the associated multi-threaded in situ analysis approach for typical production simulation of two-fluid-phase flow.

Computer Science↗

SharP Data Constructs: Data Constructs to Enable Data-Centric Computing

Extreme-scale applications (i.e., Big-Compute) are becoming increasingly data-intensive, i.e., producing and consuming increasingly large amounts of data. The HPC systems traditionally used for these applications are now used for Big-Data applications such as data analytics, social network analysis, machine learning, and genomics. As a consequence of these trends, the system architecture should be flexible and data-centric. This can already be witnessed in the pre-exascale systems with TBs of on-node hierarchical and heterogeneous memories, PBs of system memory, low-latency, high-throughput networks, and many threaded cores. As such, the pre-exascale systems suit the needs of both Big-Compute and Big-Data applications. Though the system architecture is flexible enough to support both Big-Compute and Big-Data, we argue there is a software gap. Particularly, we need data-centric abstractions to leverage the full potential of the system, i.e., there is a need for native support for data resilience, the ability to express data locality and affinity, mechanisms to reduce data movement, the ability to share data, and abstractions to express User's data usage and data access patterns. In this paper, we (i) show the need for taking a holistic approach towards data-centric abstractions, (ii) show how these approaches were realized in the SHARed data-structure centric Programming abstraction (SharP) library, a data-structure centric programming abstraction, and (iii) apply these approaches to a variety of applications that demonstrate its usefulness. Particularly, we apply these approaches to QMCPack and the Graph500 benchmark and demonstrate the advantages of this approach on extreme-scale systems.

Aderholdt, William Ferrol↗

Reconstructing the local potential of inflation with BICEP2 data

We locally reconstruct the inflationary potential by using the current constraints on r and n{sub s} from BICEP2 data. Assuming small and negligible α{sub s}, the inflationary potential is approximately linear in Δφ∼ M{sub pl} range but becomes non-linear in Δφ∼ 10 M{sub pl} range. However if we vary the value of α{sub s} within the range given by constraints from Planck measurement, the local reconstruction is only valid in the range of Δφ∼ 0.4 M{sub pl}, which challenges the inflationary background from the point of view of effective field theory. We show that, within the range of Δ φ ∼ 0.4 M{sub pl}, the inflation potential can be precisely reconstructed. With the current reconstruction, we show that V(φ) ∼ φ{sup 2} and φ{sup 3} are consistent, while φ model is ruled out by 95% confidence level of the reconstructed range of potential. This sets up a strong limit of large-field inflation models.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

On Fusing Recursive Traversals of K-d Trees

Loop fusion is a key program transformation for data locality optimization that is implemented in production compilers. But optimizing compilers currently cannot exploit fusion opportunities across a set of recursive tree traversal computations with producer-consumer relationships. In this paper, we develop a compile-time approach to dependence characterization and program transformation to enable fusion across recursively specified traversals over k-ary trees. We present the FuseT source-to-source code transformation framework to automatically generate fused composite recursive operators from an input program containing a sequence of primitive recursive operators. We use our framework to implement fused operators for MADNESS, Multiresolution Adaptive Numerical Environment for Scientific Simulation. We show that locality optimization through fusion can offer more than an order of magnitude performance improvement.

data locality↗

Performance Model of MapReduce Iterative Applications for Hybrid Cloud Bursting

Hybrid cloud bursting (i.e., leasing temporary off-premise cloud resources to boost the overall capacity during peak utilization) can be a cost-effective way to deal with the increasing complexity of big data analytics, especially for iterative applications. However, the low throughput, high latency network link between the on-premise and off-premise resources (“weak link”) makes maintaining scalability difficult. While several data locality techniques have been designed for big data bursting on hybrid clouds, their effectiveness is difficult to estimate in advance. Yet such estimations are critical, because they help users decide whether the extra pay-as-you-go cost incurred by using the off-premise resources justifies the runtime speed-up. To this end, the current paper presents a performance model and methodology to estimate the runtime of iterative MapReduce applications in a hybrid cloud-bursting scenario. The paper focuses on the overhead incurred by the weak link at fine granularity, for both the map and the reduce phases. This approach enables high estimation accuracy, as demonstrated by extensive experiments at scale using a mix of real-world iterative MapReduce applications from standard big data benchmarking suites that cover a broad spectrum of data patterns. As a result, not only are the produced estimations accurate in absolute terms compared with experimental results, but they are also up to an order of magnitude more accurate than applying state-of-art estimation approaches originally designed for single-site MapReduce deployments.

97 MATHEMATICS AND COMPUTING↗

Zoltan2

Zoltan2 is a refactoring of the open-source Zoltan library that enables Zoltan users to take better advantage of current and emerging high performance computing architectures and software tools. Like Zoltan, Zoltan2 is a toolkit of combinatorial scientific computing algorithms that enable better performance of scientific codes via improved data locality and parallel scalability. Specifically Zoltan2 partitions data, colors graphs, and orders matrices. Zoltan2 is a software library written in template C++; it supports a wider range of data types and problem sizes than Zoltan. It will be available as part of Sandia's Trilinos framework of scientific computing packages.

Riesen, Lee Ann↗

Affordability of Household Transportation Fuel Costs by Region and Socioeconomic Factors

Transportation energy is an important component of household budgets. Household vehicle fuel, one part of total transportation energy costs is over 3% of total household expenditures nationwide in the U.S. The average annual expenditure of over $2,000 for vehicle fuel (mostly gasoline) is comparable to the total average household expenditures for electricity and natural gas combined. However, these average values vary geographically, and lower income households can face higher energy cost burdens. The highly resolved data developed in this study reveals the variation of household energy burdens across the country and provides localized data to support local decision making.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

In situ compressive damage of cement paste characterized by lab source X-ray computer tomography

This paper aims at illustrating the potential of lab source X-ray CT for studying the damage behavior of cement based materials through in situ load experiments. This approach permits quantifying the microstructure prior and during loading. The load damage is separated from the specimen deformation using an image interpolation method. A quantitative relationship between external load and internal specimen damage is analyzed using the statistical information of gray scale values of the CT data. Local damage degrees are defined on 3D subset, and the 3D spatial distribution of damage information is clarified in this research. - Highlights: • On line damage is characterized by lab source X-ray CT. • Loading damage is separated with the specimen deformation. • Local damage is analyzed using gray scale values of the CT data. • 3D spatial distribution of the local damage information is clarified.

36 MATERIALS SCIENCE↗

FEDERATED LEARNING ON STOCHASTIC NEURAL NETWORKS

Federated learning is a machine learning paradigm that leverages edge computing on client devices to optimize models while maintaining user privacy by ensuring that local data remain on the device. However, since all data are collected by clients, federated learning is susceptible to latent noise in local datasets. Factors such as limited measurement capabilities or human errors may introduce inaccuracies in client data. To address this challenge, we propose the use of a stochastic neural network as the local model within the federated learning framework. Stochastic neural networks not only facilitate the estimation of the true underlying states of the data but also enable the quantification of latent noise. We refer to our federated learning approach, which incorporates stochastic neural networks as local models, as federated stochastic neural networks. In this work we will present numerical experiments demonstrating the performance and effectiveness of our method, particularly in handling nonindependent and identically distributed data.

97 MATHEMATICS AND COMPUTING↗

Federated Learning and Differential Privacy: What might AI-Enhanced co-design of microelectronics learn?

Data is a valuable commodity, and it is often dispersed over multiple entities. Sharing data or models created from the data is not simple due to concerns regarding security, privacy, ownership, and model inversion. This limitation in sharing can hinder model training and development. Federated learning can enable data or model sharing across multiple entities that control local data without having to share or exchange the data themselves. Differential privacy is a conceptual framework that brings strong mathematical guarantee for privacy protection and helps provide a quantifiable privacy guarantee to any data or models shared. The concepts of federated learning and differential privacy are introduced along with possible connections. Lastly, some open discussion topics on how federated learning and differential privacy can tied to AI-Enhanced co-design of microelectronics are highlighted.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Software-Hardware Co-design of Heterogeneous SmartNIC System for Recommendation Models Inference and Training

Deep Learning Recommendation Models (DLRMs) are critical applications in various domains and have evolved as one of the single largest machine learning applications. Trillions of DLRM parameters exceed the on-chip memory capacity of GPUs. Large-scale multi-node systems are required for distributed DLRM inference and training, which suffer from the all-to-all communication bottleneck, mainly limiting the scalability of ever-growing DLRMs. In recent years, SmartNICs have evolved with coupled computation and communication capabilities providing opportunities for a powerful heterogeneous device in the system. However, there isn't such a distributed system that fully leverages the abundant smartNIC resources that resolve the scalability issue of DLRMs. In this work, we proposed a software-hardware co-design of a heterogeneous smartNIC system that resolves the communication bottleneck of distributed DLRMs, mitigates the memory bandwidth pressure, and improves computation efficiency. We provide a set of smartNIC designs of cache systems (including local cache and remote cache) and smartNIC computation kernels which reduce data movement, relieve memory lookup intensity, and improve the GPU's computation efficiency. In addition, we propose a graph algorithm that improves the data locality of queries within batches which optimizes the overall system performance with higher data reuse. Our evaluation shows that our system achieves 2.1x latency speedup for inference and 1.6x throughput speedup for training.

Guo, Anqi↗

Develop Wake Mitigation Strategy (CRADA CRD-17-00693 Final Report)

Wind turbines in a wind farm typically operate individually to maximize their own performance and do not take into account information from nearby turbines. In an autonomous wind farm, enabling cooperation to achieve farm-level objectives, turbines will need to use information from nearby turbines to optimize performance, ensure resiliency when other sensors fail, and adapt to changing local conditions. A key element of achieving an autonomous wind farm is to develop algorithms that provide necessary information to ensure reliable, robust, and efficient operation of wind turbines in a wind plant using local sensor information that is already being collected, such as supervisory control and data acquisition (SCADA) data, local meteorological stations, and nearby radars/sodars/lidars. In this work consensus control is applied in a hybrid analysis to data from an existing wind farm to demonstrate the benefit of consensus control.

17 WIND ENERGY↗

An electrostatic Particle-In-Cell code on multi-block structured meshes

We present an electrostatic Particle-In-Cell (PIC) code on multi-block, locally structured, curvilinear meshes called Curvilinear PIC (CPIC). Multi-block meshes are essential to capture complex geometries accurately and with good mesh quality, something that would not be possible with single-block structured meshes that are often used in PIC and for which CPIC was initially developed. In spite of the structured nature of the individual blocks, multi-block meshes resemble unstructured meshes in a global sense and introduce several new challenges, such as the presence of discontinuities in the mesh properties and coordinate orientation changes across adjacent blocks, and polyjunction points where an arbitrary number of blocks meet. In CPIC, these challenges have been met by an approach that features: (1) a curvilinear formulation of the PIC method: each mesh block is mapped from the physical space, where the mesh is curvilinear and arbitrarily distorted, to the logical space, where the mesh is uniform and Cartesian on the unit cube; (2) a mimetic discretization of Poisson's equation suitable for multi-block meshes; and (3) a hybrid (logical-space position/physical-space velocity), asynchronous particle mover that mitigates the performance degradation created by the necessity to track particles as they move across blocks. The numerical accuracy of CPIC was verified using two standard plasma–material interaction tests, which demonstrate good agreement with the corresponding analytic solutions. And compared to PIC codes on unstructured meshes, which have also been used for their flexibility in handling complex geometries but whose performance suffers from issues associated with data locality and indirect data access patterns, PIC codes on multi-block structured meshes may offer the best compromise for capturing complex geometries while also maintaining solution accuracy and computational efficiency.

36 MATERIALS SCIENCE↗

Localized Actual Meteorological Year File Creator (LAF): A tool for using locally observed weather data in building energy simulations

The Localized Actual Meteorological Year File Creator (LAF) application provides web-based access to real meteorological data and processes it into a weather file suitable for building energy modeling. Building energy consumption is affected by what is inside the building (such as occupants, appliances, HVAC systems, etc.) and by what is outside: the weather conditions the building is exposed to. However, freely available weather data files are limited to a number of specific locations, usually airports, which are often located away from city centers where buildings are concentrated. The authors have developed a new tool that supports the investigation and quantification of micro-climate conditions on building energy consumption. LAF is built on the Python open-source programming language and has a Graphical User Interface (GUI) that allows users to create custom weather data files for building energy simulations. Many sets of actual meteorological year weather data for long time periods are publicly available online (such as the MesoWest database) for thousands of locations in the US. LAF selects weather data according to user specifications and automatically processes it through an API across multiple weather stations and multiple time periods. The user may easily select a specific location, time frame, and time step that best meets their needs. This article presents a useful tool for energy modelers, building designers and operators to assist with building performance analysis and optimization.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Extended experimental inferential structure determination method in determining the structural ensembles of disordered protein states

Proteins with intrinsic or unfolded state disorder comprise a new frontier in structural biology, requiring the characterization of diverse and dynamic structural ensembles. Here we introduce a comprehensive Bayesian framework, the Extended Experimental Inferential Structure Determination (X-EISD) method, which calculates the maximum log-likelihood of a disordered protein ensemble. X-EISD accounts for the uncertainties of a range of experimental data and back-calculation models from structures, including NMR chemical shifts, J-couplings, Nuclear Overhauser Effects (NOEs), paramagnetic relaxation enhancements (PREs), residual dipolar couplings (RDCs), hydrodynamic radii (R h ), single molecule fluorescence Förster resonance energy transfer (smFRET) and small angle X-ray scattering (SAXS). We apply X-EISD to the joint optimization against experimental data for the unfolded drkN SH3 domain and find that combining a local data type, such as chemical shifts or J-couplings, paired with long-ranged restraints such as NOEs, PREs or smFRET, yields structural ensembles in good agreement with all other data types if combined with representative IDP conformers.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Evaluating Wind Direction Consensus Methods: A Case Study

Wind turbines in a wind farm typically operate individually to maximize their own performance and do not take into account information from nearby turbines. To enable cooperation to achieve farm-level objectives, turbines will need to use information from nearby turbines to optimize performance, ensure resiliency when other sensors fail, and adapt to changing local conditions. A key element of achieving these cooperative wind farms is to develop algorithms that provide necessary information to ensure reliable and efficient operation of turbines in a wind plant using local sensor information that is already being collected, such as supervisory control and data acquisition (SCADA) data, local meteorological stations, and nearby radars/sodars/lidars. This presentation demonstrates a framework for implementing a cooperative wind. Farm that incorporates information from local sensors in real time to better align turbines in a wind farm. Oftentimes, measurements made at an individual turbine are noisy and unreliable. By incorporating measurements from multiple nearby turbines, a more reliable estimate of the wind direction can be obtained at an individual turbine. Results indicate that this estimate of the wind direction can be used to improve the turbine's knowledge of the wind direction and could decrease dynamic yaw misalignment, decrease the amount of time a turbines spends yawing due to a more reliable input to the yaw controller, and increase resiliency to faulty wind-vane measurements. The concept was evaluated on data provided by WindESCo and was evaluated across several other averaging techniques to demonstrate the resiliency of the proposed consensus method.

17 WIND ENERGY↗

Consensus Control (Wind Farm Collective Yaw Control) [SWR-19-11]

Wind turbines in a wind farm typically operate individually to maximize their own performance and do not take into account information from nearby turbines. In an autonomous wind farm, enabling cooperation to achieve farm-level objectives, turbines will need to use information from nearby turbines to optimize performance, ensure resiliency when other sensors fail, and adapt to changing local conditions. A key element of achieving an autonomous wind farm is to develop algorithms that provide necessary information to ensure reliable, robust, and efficient operation of wind turbines in a wind plant using local sensor information that is already being collected, such as supervisory control and data acquisition (SCADA) data, local meteorological stations, and nearby radars/sodars/lidars. This code implements a framework for implementing an autonomous wind farm that incorporates information from local sensors in real time to better align turbines in a wind farm. Oftentimes, measurements made at an individual turbine are noisy and unreliable. By incorporating measurements from multiple nearby turbines, a more robust estimate of the wind direction can be obtained at an individual turbine. Results indicate that this estimate of the wind direction can be used to improve the turbine's knowledge of the wind direction and could decrease dynamic yaw misalignment, decrease the amount of time a turbine spends yawing due to a more robust input to the yaw controller, and increase resiliency to faulty wind-vane measurements.

King, Jennifer↗