You get the pointer but only on my terms: Control local data access for better use of modern computer hardware and programming models.
Abstract not provided.
SEARCH · Search NASA
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.
Abstract not provided.
Summary Data‐driven programming models such as many‐task computing (MTC) have been prevalent for running data‐intensive scientific applications. MTC applies over‐decomposition to enable distributed scheduling. To achieve extreme scalability, MTC proposes a fully distributed task scheduling architecture that employs as many schedulers as the compute nodes to make scheduling decisions. Achieving distributed load balancing and best exploiting data locality are two important goals for the best performance of distributed scheduling of data‐intensive applications. Our previous research proposed a data‐aware work‐stealing technique to optimize both load balancing and data locality by using both dedicated and shared task ready queues in each scheduler. Tasks were organized in queues based on the input data size and location. Distributed key‐value store was applied to manage task metadata. We implemented the technique in MATRIX, a distributed MTC task execution framework. In this work, we devise an analytical suboptimal upper bound of the proposed technique, compare MATRIX with other scheduling systems, and explore the scalability of the technique at extreme scales. Results show that the technique is not only scalable but can achieve performance within 15% of the suboptimal solution. Copyright © 2015 John Wiley & Sons, Ltd.
In HPC applications, memory access behavior is one of the main factors affecting performance. Improving an application’s memory access behavior involves optimizing data layout and/or restructuring code, and requires studying spatial-temporal data locality. Existing data locality analyses focus on single-location metrics and are restricted to evaluating temporal locality. We introduce spatial-temporal affinity metrics that quantify temporal access proximity, forward access correlation, and nearby access correlation between pairs of memory locations. We describe methods for distinguishing between potential vs. realized affinity and for reasoning about affinity at multiple resolutions (3D, 2D, 1D). Finally, we construct spatial-temporal affinity signatures that classify memory behavior and that be used to reason about changes in software (data relayout, code refactoring) or hardware (caching, prefetching). We describe methods for signature visualization, interpretation, and quantitative comparison of signatures. We evaluate our methodology using applications with variants that contrast data structures, data layouts and algorithms. We show that spatial-temporal affinity analysis provides novel insights and enables predictive reasoning about application performance when contrasted with reuse distance analysis.
EERE's State and Local Energy Data (SLED) Tool provides basic energy market information that can help state and local governments plan and implement clean energy projects, including electricity generation; fuel sources and costs; applicable policies, regulations, and financial incentives; and renewable energy resource potential. Watch this video to learn more about the tool and hear testimonials from real users about the benefits of using this tool.
A new parallel dynamic mesh adaptation and load balancing algorithm for general hybrid grids has been developed. The meshes considered in this work are composed of four kinds of elements; tetrahedra, prisms, hexahedra and pyramids, which poses a challenge to parallel mesh adaptation. Additional complexity imposed by the presence of multiple types of elements affects especially data migration, updates of local data structures and interpartition data structures. Efficient partition of hybrid meshes has been accomplished by transforming them to suitable graphs and using serial graph partitioning algorithms. Communication among processors is based on the faces of the interpartition boundary and the termination detection algorithm of Dijkstra is employed to ensure proper flagging of edges for refinement. An inexpensive dynamic load balancing strategy is introduced to redistribute work load among processors after adaptation. In particular, only the initial coarse mesh, with proper weighting, is balanced which yields savings in computation time and relatively simple implementation of mesh quality preservation rules, while facilitating coarsening of refined elements. Special algorithms are employed for (i) data migration and dynamic updates of the local data structures, (ii) determination of the resulting interpartition boundary and (iii) identification of the communication pattern of processors. Several representative applications are included to evaluate the method.
Recorded seismic data are generally contaminated by noise from different sources, which masks the signals of interest. In the seismology community, frequency filtering (FF) is the standard method for noise suppression. However, when the signal of interest and noise share the same frequency band, the latter cannot be filtered out without infringing on the former. We implemented a noise suppression approach based on the mathematical morphology theorem. The method involves compound operations of dilation and erosion using structuring elements of varying lengths and decomposes an input noisy waveform into several time functions with differing characteristics. Further, the filtered waveform is constructed from the time functions using a self-adaptive reconstruction technique. Application to a data set of >4700 local waveforms suggests that the implemented mathematical morphological filtering (MMF) approach is efficient for data with low signal-to-noise ratio (SNR) and significantly outperforms FF in that SNR range. For most of the dataset, FF, machine learning (ML) denoising, and continuous wavelet transform (CWT) thresholding result in higher SNR values compared with the MMF method. However, for ~42% of the waveforms, MMF outperforms FF, and the SNR gain achieved with MMF is as large as ~23 dB. Compared to ML denoising and CWT thresholding, this proportion drops to only ~10%–14%. Our results suggests that in an operational setting, MMF cannot replace the other noise suppression methods; however, signal detection can be improved if MMF is used to supplement them in some scenarios. MMF could help detect signals in problematic low-SNR data, which are currently being missed particularly when using FF alone.
Simulations of core-collapse supernovae, and other astrophysical phenomena, are quintessential extreme-scale computing challenges. For core-collapse supernova simulations to be carried out by the ExaStar project under the Exascale Computing Project umbrella, a robust, efficient, and state-of-the-art magnetohydrodynamics solver is a critical requirement. In Flash-X, the primary software instrument for ExaStar, a new magnetohydrodynamics solver has been designed and implemented from the ground up to achieve accuracy and efficiency for simulations of complex astrophysical flows. This new solver, dubbed Spark, uses high-order spatial reconstruction, Runge-Kutta time integration, and an efficient cell-centered approach to satisfying the divergence-free condition for the magnetic fields. Spark was written to be optimized for data locality in cache hierarchy of CPUs. Since data locality optimizations for cache hierarchy are not directly compatible with those of accelerators, we have taken the approach of using program synthesis to avoid massive amounts of code replication that would be necessary if we were to maintain two different versions of the solver. Our program synthesis relies on a simple key-dictionary approach, implemented in python, that enables us to assemble the version of the solver suitable for the target hardware from code fragments identified by specific keys. In this work, we describe the data locality optimizations of the solver for CPUs and accelerators and the program synthesis tools that enable this portability. We also detail the parallel performance of Spark for both CPUs and accelerators.
This study evaluated how applicable European Life Cycle Inventory (LCI) data are to assessing the environmental impacts of the life cycle of Brazilian triple superphosphate (TSP). The LCI data used for the comparison were local Brazilian LCI data, European LCI data in its original version from the ecoinvent database and a modified version of the European LCI data, which had been adapted to better account for the Brazilian situation. We compared the three established datasets at the level of the inventory as well as for their environmental impacts, i.e. at the level of Life Cycle Environmental Assessment (LCIA). The analysis showed that the European LCIs (both the original and the modified ones) considered a broader spectrum of background processes and environmental flows (inputs and outputs). Nevertheless, TSP production had in all three cases similar values for the consumption of the main raw materials. The LCIA results obtained for the datasets showed important differences as well. Therefore we concluded that the European data in general lead to much higher environmental impacts than the Brazilian data. The differences between the LCIA results obtained with the Brazilian and the European data can be basically explained by the methodological differences underlying the data. The small differences at the LCI level for selected inputs and outputs between the Brazilian and the European LCIs from ecoinvent indicate that the latter can be regarded as applicable for characterizing the Brazilian TSP.
Local ordering of atoms within a crystalline structure is increasingly recognized as essential to understanding the properties of a variety of interesting materials. This is shown by the success of pair distribution function (PDF) analysis of scattering data from powder samples to investigate departures from perfectly crystalline order (Billinge & Kanatzidis, 2004; Egami & Billinge, 2012). Diffuse scattering from single crystals can also be a powerful probe of local order, potentially providing key information that would be difficult to obtain from PDF methods; however, difficulties in both measurement and interpretation of data have too often restricted their use by a larger community. In this issue of IUCrJ, Rothet al.(2020) demonstrate how simple concepts can be applied to better understand single-crystal diffuse scattering from half-Heusler systems.
This presentation provides an overview of the improvement of aerosol optical depth data for localized solar forecasting.
Measuring changes in protein or organelle abundance in the cell is an essential, but challenging aspect of cell biology. Frequently-used methods for determining organelle abundance typically rely on detection of a very few marker proteins, so are unsatisfactory. In silico estimates of protein abundances from publicly available protein spectra can provide useful standard abundance values but contain only data from tissue proteomes, and are not coupled to organelle localization data. A new protein abundance score, the normalized protein abundance scale (NPAS), expands on the number of scored proteins and the scoring accuracy of lower-abundance proteins in Arabidopsis. NPAS was combined with subcellular protein localization data, facilitating quantitative estimations of organelle abundance during routine experimental procedures. A suite of targeted proteomics markers for subcellular compartment markers was developed, enabling independent verification of in silico estimates for relative organelle abundance. Estimation of relative organelle abundance was found to be reproducible and consistent over a range of tissues and growth conditions. In silico abundance estimations and localization data have been combined into an online tool, multiple marker abundance profiling, available in the SUBA4 toolbox (http://suba.live).
As the seismic monitoring community advances toward detecting, identifying, and locating ever-smaller natural and anthropogenic events, the need is constantly increasing for higher resolution, higher fidelity data, models, and methods for accurately characterizing events. Local-distance seismic data provide robust constraints on event locations, but also introduce complexity due to the significant geologic heterogeneity of the Earth’s crust and upper mantle, and the relative sparsity of data that often occurs with small events recorded on regional seismic networks. Identifying the critical characteristics for improving local-scale event locations and the factors that impact location accuracy and reliability is an ongoing challenge for the seismic community. Using Utah as a test case, we examine three data sets of varying duration, finesse, and magnitude to investigate the effects of local earth structure and modeling parameters on local-distance event location precision and accuracy. We observe that the most critical elements controlling relocation precision are azimuthal coverage and local-scale velocity structure, with tradeoffs based on event depth, type, location, and range.
Big data processing systems such as Spark are employed in an increasing number of diverse applications-such as machine learning, graph computation, and scientific computing-each with dynamic and different resource needs. These applications increasingly run on heterogeneous hardware, e.g., with out-of-core accelerators. However, big data platforms do not factor in the multi-dimensional heterogeneity of applications and hardware. This leads to a fundamental mismatch between the application and hardware characteristics, and the resource scheduling adopted in big data platforms. For example, Hadoop and Spark consider only data locality when assigning tasks to nodes, and typically disregard the hardware capabilities and suitability to specific application requirements. In this paper, we present RUPAM, a heterogeneity-aware task scheduling system for big data platforms, which considers both task-level resource characteristics and underlying hardware characteristics, as well as preserves data locality. RUPAM adopts a simple yet effective heuristic to decide the dominant scheduling factor (e.g., CPU, memory, or I/O), given a task in a particular stage. Our experiments show that RUPAM is able to improve the performance of representative applications by up to 62.3% compared to the standard Spark scheduler.
This fact sheet summarizes some of the key needs, barriers, and opportunities related to data center management and operations in the state and local government sector.
Large Language Models have proven adept at answering general knowledge questions. To make these generative AI tools useful to our mission customers for monitoring global threats, the data sciences team at Sandia is utilizing retrieval augmented generation (RAG) techniques to customize these models with local data. The local data we use consists of data such as research articles and patent abstracts that we've collected over the last several years using automated pipelines.
CAP-88 PC is a commonly used radiological atmospheric dispersion model. This US EPA-approved model is used to demonstrate compliance with atmospheric emission regulations for radionuclides. While the model includes a large library of meteorological data for use across the United States, there are applications when users may wish to use onsite meteorological data as an input to the CAP-88 PC model. Here, in this work, we present a work-around process for preparing and converting onsite data for use in CAP-88 PC. However, the use of local data should provide a more realistic estimate of doses to members of the public in the immediate vicinity of a facility, although the regulatory agency having jurisdiction may not accept the use of local data for compliance. Additionally, the historical meteorological records from 20+ years ago (at a site many kilometers away) might not be representative of current local weather patterns, highlighting another benefit of using local meteorological data.
Modern experiments are typically conducted by large, extended groups, where researchers rely on other team members to produce much of the data they use. The experiments record very large numbers of measurements that can be difficult for users to find, access and understand. Here. we are developing a system for users to annotate their data products with structured metadata, providing data consumers with a discoverable, browsable data index. Machine understandable metadata captures the underlying semantics of the recorded data, which can then be consumed by both programs, and interactively by users. Collaborators can use these metadata to select and understand recorded measurements. The data catalog project is a data dictionary and index which enables users to record general descriptive metadata, use cases and rendering information as well as providing them a transparent data access mechanism (URI). Users describe their diagnostic including references, text descriptions, units, labels, example data instances, author contact information and data access URIs. The list of possible attribute labels is extensible, but limiting the vocabulary of names increases the utility of the system. The data catalog is focused on the data products and complements process-based systems like the Metadata Ontology Provenance project [Greenwald, 2012; Schissel, 2015]. This system can be coupled with MDSplus to provide a simple platform for data driven display and analysis programs. Sites which use MDSplus can describe tree branches, and if desired create ‘processed data trees’ with homogeneous node structures for measurements. Sites not currently using MDSplus can either use the database to reference local data stores, or construct an MDSplus tree whose leaves reference the local data store. A data catalog system can provide a useful roadmap of data acquired from experiments or simulations making it easier for researchers to find and access important data and understand the meaning of the data and how it was obtained. This is particularly useful in research facilities that study the results of many different experiments or simulations and may not know the intricacies of the data organization in use where the data was generated. It is also possible to store a local copy of key data items in local MDSplus trees and then add processed data to the local catalog.
In this work, we study the network locality improvements that can be achieved by using co-packaged optics in data center and high-performance computing (HPC) networks. The increased escape bandwidth offered by co-packaged optics can enable switches with speeds of 51.2 Tb/s and beyond. From a network architecture perspective, the key advantages of introducing co-packaged optics at the switch points include the implementation of large-scale topologies of >12,000 end points with 4× higher bisection bandwidth and the reduction of the required number of switches by >40% compared with state-of-the-art approaches. From a network operation perspective, improved network locality and faster operation can be achieved since the higher-radix switches can mitigate the impact of network contention. Placing applications under fewer leaf switches reduces the number of packets that cross the spine switches in a leaf-spine topology. The proposed scheme is evaluated via discrete-event simulations: we initially evaluate the network locality properties of the system by using virtual-machine traces from a production data center, and we subsequently quantify the performance improvements by simulating an all-to-all pattern for a variety of message sizes over a number of nodes. The results suggest that co-packaged optics form a promising solution for keeping up with bandwidth scaling in future networks. The virtual-machine analysis shows that large-scale applications can be placed under up to 50% fewer first-level switches, while the network analysis shows speedups of up to 7.1, which translates to execution time reductions of up to 26% and 42.7% for applications with communication ratios of 0.3 and 0.5, respectively.