Publications

Abstract not provided.

Abstract not provided.

We present a resilient domain-decomposition preconditioner for partial differential equations (PDEs). The algorithm reformulates the PDE as a sampling problem, followed by a solution update through data manipulation that is resilient to both soft and hard faults. We discuss an implementation based on a server-client model where all state information is held by the servers, while clients are designed solely as computational units. Servers are assumed to be “sandboxed”, while no assumption is made on the reliability of the clients. We explore the scalability of the algorithm up to ∼12k cores, build an SST/macro skeleton to extrapolate to∼50k cores, and show the resilience under simulated hard and soft faults for a 2D linear Poisson equation.

Abstract not provided.

In this work we have presented a particle resuspension model implemented in the SNL code SIERRA/Fuego, which can be used to model particle dispersal and resuspension from surfaces. The method demonstrated is applicable to a class of particles, but would require additional parametric fits or physics models for extension to other applications, such as wetted particles or walls. We have demonstrated the importance of turbulent variations in the wall shear stress when considering resuspension, and implemented both shear stress variation models and stochastic resuspension models (not shown in this work). These models can be used in simulations with of physically realistic scenarios to augment lab-scale DOE Handbook data for airborne release fractions and respirable fractions in order to provide confidences for safety analysts and facility designers to apply in their analyses at DOE sites. Future work on this topic will involve validation of the presented model against experimental data and extension of the empirical models to be applicable to different classes of particles and surfaces.

Recent advances in sensor technology have made continuous real-time health monitoring available in both hospital and non-hospital settings. Since data collected from high frequency medical sensors includes a huge amount of data, storing and processing continuous medical data is an emerging big data area. Especially detecting anomaly in real time is important for patients' emergency detection and prevention. A time series discord indicates a subsequence that has the maximum difference to the rest of the time series subsequences, meaning that it has abnormal or unusual data trends. In this study, we implemented two versions of time series discord detection algorithms on a high performance parallel database management system (DBMS) and applied them to 240 Hz waveform data collected from 9,723 patients. The initial brute force version of the discord detection algorithm takes each possible subsequence and calculates a distance to the nearest non-self match to find the biggest discords in time series. For the heuristic version of the algorithm, a combination of an array and a trie structure was applied to order time series data for enhancing time efficiency. The study results showed efficient data loading, decoding and discord searches in a large amount of data, benefiting from the time series discord detection algorithm and the architectural characteristics of the parallel DBMS including data compression, data pipe-lining, and task scheduling.

This work quantifies the polarization persistence and memory of circularly polarized light in forward-scattering and isotropic (Rayleigh regime) environments; and for the first time, details the evolution of both circularly and linearly polarized states through scattering environments. Circularly polarized light persists through a larger number of scattering events longer than linearly polarized light for all forward-scattering environments; but not for scattering in the Rayleigh regime. Circular polarization's increased persistence occurs for both forward and backscattered light. The simulated environments model polystyrene microspheres in water with particle diameters of 0.1 μm, 2.0 μm, and 3.0 μm. The evolution of the polarization states as they scatter throughout the various environments are illustrated on the Poincaré sphere after one, two, and ten scattering events.

While public-key cryptography is essential for secure communications, the energy cost of even the most efficient algorithms based on Elliptic Curve Cryptography (ECC) is prohibitive on many ultra-low energy devices such as sensornetwork nodes and identification tags. Although an abundance of hardware acceleration techniques for ECC have been proposed in literature, little research has focused on understanding the energy benefits of these techniques. Therefore, we evaluate the energy cost of ECC on several different hardware/software configurations across a range of security levels. Our work comprehensively explores implementations of both GF(p) and GF(2m) ECC, demonstrating that GF(2m) provides a 1.31 to 2.11 factor improvement in energy efficiency over GF(p) on an extended RISC processor. We also show that including a 4KB instruction cache in our system can reduce the energy cost of ECC by as much as 30%. Furthermore, our GF(2m) coprocessor achieves a 2.8 to 3.61 factor improvement in energy efficiency compared to instruction set extensions and significantly outperforms prior work.

In situ X-Ray Absorption Near Edge Spectroscopy (XANES) and Extended X-Ray Absorption Fine Structure (EXAFS) techniques are applied to a metal center ionic liquid undergoing oxidation and reduction in a three electrode spectroscopic cell. Determination of the extent of reduction under negative bias on the working electrode and the extent of oxidation are determined after pulse voltammetry to quiescence. While the ionic liquid undergoes full oxidation, it undergoes only partial reduction, likely due to transport issues on the timescale of the experiment. Nearest neighbor Fe-O distances in the fully oxidized state match well to expected values for similarly coordinated solids, but reduction does not result in an extension of the Fe-O bond length, as would be expected from comparisons to the solid phase. Instead, little change in bond length is observed. We suggest that this may be due to a more complex interaction between the monodentate ligands of the metal center anion and the surrounding charge cloud, rather than straightforward electrostatics between the metal center and the nearest neighbor grouping.

Safety basis analysts throughout the U.S. Department of Energy (DOE) complex rely heavily on the information provided in the DOE Hand book, DOE-HDBK-3010, Airborne Release Fractions/Rates and Resp irable Fractions for Nonreactor Nuclear Facilities , to determine source terms. In calcula ting source terms, analysts tend to use the DOE Handbook's bounding values on airbor ne release fractions (ARFs) and respirable fractions (RFs) for various cat egories of insults (representing potential accident release categories). This is typica lly due to both time constraints and the avoidance of regulatory critique. Unfort unately, these bounding ARFs/RFs represent extremely conservative values. Moreover, th ey were derived from very limited small- scale table-top and bench/labo ratory experiments and/or fr om engineered judgment. Thus the basis for the data may not be re presentative to the actual unique accident conditions and configura tions being evaluated. The goal of this res earch is to develop a more ac curate method to identify bounding values for the DOE Handbook using the st ate-of-art multi-physics-based high performance computer codes. This enable s us to better understand the fundamental physics and phenomena associated with the ty pes of accidents for the data described in it. This research has examined two of the DOE Handbook's liquid fire experiments to substantiate the airborne release frac tion data. We found th at additional physical phenomena (i.e., resuspension) need to be included to derive bounding values. For the specific cases of solid powder under pre ssurized condition and mechanical insult conditions the codes demonstrated that we can simulate the phenomena. This work thus provides a low-cost method to establis h physics-justified sa fety bounds by taking into account specific geometri es and conditions that may not have been previously measured and/or are too costly to do so.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Soil from an excavated test was being surveyed and fresh DU fragments removed so the soil could be returned to the excavation hole. Sandia RCTs discovered two locations where highly oxidized DU was present just below the undisturbed lake bed surface. The oxidized DU was removed for disposal as radioactive waste. Approximately one cubic foot of oxidized DU mixed with soil was removed. Line Manager, Sr. Manager and Acting Director were all notified of this event.

Power consumption of extreme-scale supercomputers has become a key performance bottleneck. Yet current practices do not leverage power management opportunities, instead running at maximum power. This is not sustainable. Future systems will need to manage power as a critical resource, directing it to where it has greatest benefit. Power capping is one mechanism for managing power budgets, however its behavior is not well understood. This paper presents an empirical evaluation of several key HPC workloads running under a power cap on a Cray XC40 system, and provides a comparison of this technique with p-state control, demonstrating the performance differences of each. These results show: 1.) Maximum performance requires ensuring the cap is not reached; 2.) Performance slowdown under a cap can be attributed to cascading delays which result in unsynchronized performance variability across nodes; and, 3.) Due to lag in reaction time, considerable time is spent operating above the set cap. This work provides a timely and much needed comparison of HPC application performance under a power cap and attempts to enable users and system administrators to understand how to best optimize application performance on power-constrained HPC systems.

The moon-forming impact and the subsequent evolution of the proto-Earth is strongly dependent on the properties of materials at the extreme conditions generated by this violent collision. We examine the high pressure behavior of MgO, one of the dominant constituents in Earth's mantle, using high-precision, plate impact shock compression experiments performed on Sandia National Laboratories' Z Machine and extensive quantum calculations using density functional theory (DFT) and quantum Monte Carlo (QMC) methods. The combined data span from ambient conditions to 1.2 TPa and 42 000 K, showing solid-solid and solid-liquid phase boundaries. Furthermore our results indicate that under impact the solid and liquid phases coexist for more than 100 GPa, pushing complete melting to pressures in excess of 600 GPa. The high pressure required for complete shock melting has implications for a broad range of planetary collision events.

Abstract not provided.

Abstract not provided.