Publications

SciDAC applications have a demonstrated need for advanced software tools to manage the complexities associated with sophisticated geometry, mesh, and field manipulation tasks, particularly as computer architectures move toward the petascale. In this paper, we describe a software component - an abstract data model and programming interface - designed to provide support for parallel unstructured mesh operations. We describe key issues that must be addressed to successfully provide high-performance, distributed-memory unstructured mesh services and highlight some recent research accomplishments in developing new load balancing and MPI-based communication libraries appropriate for leadership class computing. Finally, we give examples of the use of parallel adaptive mesh modification in two SciDAC applications. © 2009 IOP Publishing Ltd.

There are two principal scientific objectives in the study of Type Ia supernovae - first, a better understanding of these complex explosions from as near first principles as possible, and second, enabling the more accurate utilization of their emission to measure distances in cosmology. Both tasks lend themselves to large scale numerical simulation, yet take us beyond the current frontiers in astrophysics, combustion science, and radiation transport. Their study requires novel approaches and the creation of new, highly scalable codes. © 2009 IOP Publishing Ltd.

Abstract not provided.

An experiment using fins mounted on a wind tunnel wall has examined the proposition that the interaction between axially-separated aerodynamic control surfaces fundamentally results from an angle of attack superposed upon the downstream fin by the vortex shed from the upstream fin. Particle Image Velocimetry data captured on the surface of a single fin show the formation of the trailing vortex first as a leading-edge vortex, then becoming a tip vortex as it propagates to the fin's spanwise edge. Data acquired on the downstream fin surface in the presence of a trailing vortex shed from an upstream fin may remove this impinging vortex by subtracting its mean velocity field as measured in single-fin experiments, after which the vortex forming on the downstream fin's leeside becomes evident. The properties of the downstream fin's lifting vortex appear to be determined by the total angle of attack imposed upon it, which is a combination of its physical fin cant and the angle of attack induced by the impinging vortex, and are consistent with those of a single fin at equivalent angle of attack.

This paper describes an application driven methodology for understanding the impact of future architecture decisions on the end of the MPP era. Fundamental transistor device limitations combined with application performance characteristics have created the switch to multicore/multithreaded architectures. Designing large-scale supercomputers to match application demands is particularly challenging since performance characteristics are highly counter-intuitive. In fact, data movement more than FLOPS dominates. This work discusses some basic performance analysis for a set of DOE applications, the limits of CMOS technology, and the impact of both on future architectures. © 2009 IOP Publishing Ltd.

We have developed algorithms for a virtual presence and extended defense (VPED) system that automatically learns the detection map of a deployed sensor field without a-priori knowledge of the local terrain. The VPED system is a network of sensor pods, with each pod containing acoustic and seismic sensors. Each pod has a limited detection range, but a network of pods can form a virtual perimeter. The site's geography and soil conditions can affect the detection performance of the pods. Thus a network in the field may not have the same performance as a network designed in the lab. To solve this problem we automatically estimate a network's detection performance as it is being constructed. We demonstrate results using simulated and real data. © 2009 IEEE.

While connectivity is an important aspect of heterogeneous media, methods to measure and simulate connectivity are limited. For this study, we use natural aquifer analogs developed through lidar imagery to track the importance of connectivity on dispersion characteristics. A 221.8 cm by 50 cm section of a braided sand and gravel deposit of the Ceja Formation in Bernalillo County, New Mexico is selected for the study. The use of two-point (SISIM) and multipoint (Snesim and Filtersim) stochastic simulation methods are then compared based on their ability to replicate dispersion characteristics using the aquifer analog. Detailed particle tracking simulations are used to explore the streamline-based connectivity that is preserved using each method. Connectivity analysis suggests a strong relationship between the length distribution of sand and gravel facies along streamlines and dispersion characteristics.

Process Systems Engineering (PSE) is built on the application of computational tools to the solution of physical engineering problems. Over the course of its nearly five decade history, advances in PSE have relied roughly equally on advancements in desktop computing technology and developments of new tools and approaches for representing and solving problems (Westerberg, 2004). Just as desktop computing development over that period focused on increasing the net serial instruction rate, tool development in PSE has emphasized creating faster general-purpose serial algorithms. However, in recent years the increase in net serial instruction rate has slowed dramatically, with processors first reaching an effective upper limit for clock speed and now approaching apparent limits for microarchitecture efficiency. Current trends in desktop processor development suggest that future performance gains will occur primarily through exploitation of parallelism. For PSE to continue to leverage the "free" advancements from desktop computing technology in the future, the PSE toolset will need to embrace the use of parallelization. Unfortunately, "parallelization" is more than just identifying multiple things to do at once. Parallel algorithm design has two fundamental challenges: first, to match the characteristics of the parallelizable problem workload to the capabilities of the hardware platform, and second to properly balance parallel computation with the overhead of communication and synchronization on that platform. The performance of any parallel algorithm is thus a strong function of how well the characteristics of the problem and algorithm match those of the hardware platform on which it will run. This has led to a proliferation of highly specialized parallel hardware platforms, each designed around specific problems or problem classes. While every platform has its own unique characteristics, we can group current approaches into six basic classes: symmetric multiprocessing (SMP), networks of workstations (NOW), massively parallel processing (MPP), specialized coprocessors, multi-threaded shared memory, and hybrids that combine components of the first five classes. Perhaps the most familiar of these is the SMP architecture, which forms the bulk of current the desktop and workstation market. These systems have multiple processing units (processors and/or cores) controlled by a single operating system image and sharing a single common shared memory space. While SMP systems provide only a modest level of parallelism (typically 2-16 processing units), the existence of shared memory and full-featured processing units makes them perhaps the most straightforward development platform. A challenge of SMP platforms is the discrepancy between the speed of the processor and the memory system: both latency and overall memory bandwidth limitations can lead to processors idling waiting for data. Clusters, a generic term for coordinated groups of independent computers (nodes) connected with high-speed networks, provide the opportunity for a radically different level of parallelism, with the largest clusters having over 25,000 nodes and 100,000 processing units. The challenge with clusters is memory is distributed across independent nodes. Communication and coordination among nodes must be explicitly managed and occurs over a relatively high latency network interconnect. Efficient use of this architecture requires applications that decompose into pseudo-independent components that run with high computation to communication ratios. The level to which systems utilize commodity components distinguishes the two main types of cluster architectures, with NOW nodes running commodity network interconnects and operating systems and MPP nodes using specialized or proprietary network layers or microkernels. Specialized coprocessors, including graphics processing units (GPU) and the Cell Broadband Engine (Cell), are gaining popularity as scientific computing platforms. These platforms employ non-general purpose dependent processing units to speed fine-grained, repetitive processing. Architecturally, they are reminiscent of vector computing, combining very fast access to a small amount of local memory with processing elements implementing either a single-instruction-multiple-data (SIMD) (GPU) or a pipelined (Cell) model. As application developers must explicitly manage both parallelism on the coprocessor and the movement of data to and from the coprocessor memory space, these architectures can be some of the most challenging to program. Finally, multi-threaded shared memory (MTSM) systems represent a fundamental departure from traditional distributed memory systems like NOW and MPP. Instead of a collection of independent nodes and memory spaces, an MTSM system runs a single system image across all nodes, combining all node memory into a single coherent shared memory space. To a developer, the MTSM appears to be a single very large SMP. However, unlike a SMP that uses caches to reduce the latency of a memory access, the MTSM tolerates latency by using a large number of concurrent threads. While this architecture lends itself to problems that are not readily decomposable, effective utilization of MTSM systems requires applications to run hundreds - or thousands - of concurrent threads. The proliferation of specialized parallel computing architectures presents several significant challenges for developers of parallel modeling and optimization applications. Foremost is the challenge of selecting the "appropriate" platform to target when developing the application. While it is clear that architectural characteristics can significantly affect the performance of an algorithm, relatively few rules or heuristics exist for selecting a platform based solely on application characteristics. A contributing challenge is that different architectures employ fundamentally different programming paradigms, libraries, and tools. Knowledge and experience on one platform does not necessarily translate to other platforms. This also complicates the process of directly comparing platform performance, as applications are rarely portable: software designed for one platform rarely compiles on another without modification, and the modifications may require a redesign of the fundamental parallelization approach. A final challenge is effectively communicating parallel results. While the relatively homogenous environment of serial desktop computing facilitated extremely terse descriptions of a test platform, often limited to processor make and clock speed, reporting results for parallel architectures must include not only processor information, but depending on the architecture, also include operating system, network interconnect, coprocessor make, model, and interconnect, and node configuration. There are numerous examples of algorithms and applications designed explicitly to leverage specific architectural features of parallel systems. While by no means comprehensive, three current representative efforts are the development of parallel branch and bound algorithms, distributed collaborative optimization algorithms, and multithreaded parallel discrete event simulation. PICO, the Parallel Integer and Combinatorial Optimizer (Eckstein, et al., 2001), is a scalable parallel mixed-integer linear optimizer. Designed explicitly for cluster environments (both NOW and MPP), PICO leverages the synergy between the inherently decomposable branch and bound tree search and the independent nature of the nodes within a cluster by distributing the independent sub-problems for the tree search across the nodes of the cluster. In contrast, agent-based collaborative optimization (Siirola, et al., 2004, 2007) matches traditionally non-decomposable nonlinear programming algorithms to high-latency clusters (e.g. NOWs or Grids) by replicating serial search algorithms intact and unmodified across the independent nodes of the cluster. The system then enforces collaboration through sharing intermediate "solutions" to the common problem. This creates a decomposable artificial meta-algorithm with a high computation to communication ratio that can scale efficiently on large, high latency, low bandwidth cluster environments. For modeling applications, efficiently parallelizing discrete event simulations has presented a longstanding challenge, with several decades of study and literature (Perumalla, 2006). The central challenge in parallelizing discrete event simulations on traditional distributed memory clusters is efficiently synchronizing the simulation time across the processing nodes during a simulation. A promising alternative approach leverages the Cray XMT (formerly called Eldorado; Feo, et al. 2005). The XMT implements an MTSM architecture and provides a single shared memory space across all nodes, greatly simplifying the time synchronization challenge. Further, the fine-grained parallelism provided by the architecture opens new opportunities for additional parallelism beyond simple event parallelization, for example, parallelizing the event queue management. While these three examples are a small subset of current parallel algorithm design, they demonstrate the impact that parallel architectures have had and will continue to have on future developments for modeling and optimization in PSE. © 2009 Elsevier B.V. All rights reserved.

Widely applied to RF filtering, AlN microresonators offer the ability to perform additional functions such as impedance matching and single-ended-to- differential conversion. This paper reports microresonators capable of transforming the characteristic impedance from input to output over a wide range while performing low loss filtering. Microresonant transformer theory of operation and equivalent circuit models are presented and compared with measured 2 and 3-Port devices. Impedance transformation ratios as large as 18:1 are realized with insertion losses less than 5.8 dB, limited by parasitic shunt capacitance. These impedance transformers occupy less than 0.052 mm2, orders of magnitude smaller than competing technologies in the VHF and UHF frequency bands. ©2009 IEEE.

Popular nuclear spectral analysis applications typically use either the results of a peak search or of the best match of a set of linear templates as the basis for their conclusions. These well-proven methods work well in controlled environments. However, they often fail in cases where the critical information resides in well-masked peaks, where the data is sparse and good statistics cannot be obtained, and where little is known about the detector that was used. These conditions are common in emergency analysis situations, but are also common in radio-assay situations where background radiation is high and time is limited. To address these limitations, non-linear fitting techniques have been introduced into an application called ''Cambio'' suitable for public use. With this approach, free parameters are varied in iterative steps to converge to values that minimize differences between the actual data and the approximating functions that correspond to the values of the parameters. For each trial nuclide, a single parameter is varied that often has a strongly non-linear dependence on other, simultaneously varied parameters for energy calibration, attenuation by intervening matter, detector resolution, and peak-shape deviations. A brief overview of this technique and its implementation is presented, together with an example of its performance and differences from more common methods of nuclear spectral analysis. © Akadémiai Kiadó, 2009.

In this paper we describe Megatux, a set of tools we are developing for rapid provisioning of millions of virtual machines and controlling and monitoring them, as well as what we've learned from booting one million Linux virtual machines on the Thunderbird (4660 nodes) and 550,000 Linux virtual machines on the Hyperion (1024 nodes) clusters. As might be expected, our tools use hierarchical structures. In contrast to existing HPC systems, our tools do not require perfect hardware; that all systems be booted at the same time; and static configuration files that define the role of each node. While we believe these tools will be useful for future HPC systems, we are using them today to construct botnets. Botnets have been in the news recently, as discoveries of their scale (millions of infected machines for even a single botnet) and their reach (global) and their impact on organizations (devastating in financial costs and time lost to recovery) have become more apparent. A distinguishing feature of botnets is their emergent behavior: fairly simple operational rule sets can result in behavior that cannot be predicted. In general, there is no reducible understanding of how a large network will behave ahead of 'running it'. 'Running it' means observing the actual network in operation or simulating/emulating it. Unfortunately, this behavior is only seen at scale, i.e. when at minimum 10s of thousands of machines are infected. To add to the problem, botnets typically change at least 11% of the machines they are using in any given week, and this changing population is an integral part of their behavior. The use of virtual machines to assist in the forensics of malware is not new to the cyber security world. Reverse engineering techniques often use virtual machines in combination with code debuggers. Nevertheless, this task largely remains a manual process to get past code obfuscation and is inherently slow. As part of our cyber security work at Sandia National Laboratories, we are striving to understand the global network behavior of botnets. We are planning to take existing botnets, as found in the wild, and run them on HPC systems. We have turned to HPC systems to support the creation and operation of millions of Linux virtual machines as a means of observing the interaction of the botnet and other noninfected hosts. We started out using traditional HPC tools, but these tools are designed for a much smaller scale, typically topping out at one to ten thousand machines. HPC programming libraries and tools also assume complete connectivity between all nodes, with the attendant configuration files and data structures to match; this assumption holds up very poorly on systems with millions of nodes.

This paper presents a new nonlinear control methodology for slewing spacecraft, which provides both necessary and sufficient conditions for stability by identifying the stability boundaries, rigid body modes, and limit cycles. Conservative Hamiltonian system concepts, which are equivalent to static stability of airplanes, are used to find and deal with the static stability boundaries: rigid body modes. The application of exergy and entropy thermodynamic concepts to the work-rate principle provides a natural partitioning through the second law of thermodynamics of power flows into exergy generator, dissipator, and storage for Hamiltonian systems that is employed to find the dynamic stability boundaries: limit cycles. This partitioning process enables the control system designer to directly evaluate and enhance the stability and performance of the system by balancing the power flowing into versus the power dissipated within the system subject to the Hamiltonian surface (power storage). Relationships are developed between exergy, power flow, static and dynamic stability, and Lyapunov analysis. The methodology is demonstrated with two illustrative examples: (1) a nonlinear oscillator with sinusoidal damping and (2) a multi-input-multioutput three-axis slewing spacecraft that employs proportional-integral-derivative tracking control with numerical simulation results.

Research on the effects of collaboration in scientific research has been increasing in recent years. A variety of studies have been done at the institution and country level, many with an eye toward policy implications. However, the question of how to identify the most fruitful targets for future collaboration in high-performing areas of science has not been addressed. This paper presents a method for identifying targets for future collaboration between two institutions. The utility of the method is shown in two different applications: identifying specific potential collaborations at the author level between two institutions, and generating an index that can be used for strategic planning purposes. Identification of these potential collaborations is based on finding authors that belong to the same small paper-level community (or cluster of papers), using a map of science and technology containing nearly 1 million papers organized into 117,435 communities. The map used here is also unique in that it is the first map to combine the ISI Proceedings database with the Science and Social Science Indexes at the paper level. © 2008 Springer Science+Business Media B.V.

To date, research in trust negotiation has focused mainly on the theoretical aspects of the trust negotiation process, and the development of proof of concept implementations. These theoretical works and proofs of concept have been quite successful from a research perspective, and thus researchers must now begin to address the systems constraints that act as barriers to the deployment of these systems. To this end, we present TrustBuilder2, a fully-configurable and extensible framework for prototyping and evaluating trust negotiation systems. TrustBuilder2 leverages a plug-in based architecture, extensible data type hierarchy, and flexible communication protocol to provide a framework within which numerous trust negotiation protocols and system configurations can be quantitatively analyzed. In this paper, we discuss the design and implementation of TrustBuilder2, study its performance, examine the costs associated with flexible authorization systems, and leverage this knowledge to identify potential topics for future research, as well as a novel method for attacking trust negotiation systems.

Polycrystalline diamond compact (PDC) bits have gained i wide popularity in the petroleum industry for drilling soft and; moderately firm formations. However, in hard formation applications, the PDC bit still has limitations, even though recent developments in PDC cutter designs and materials steadily imj proves PDC bit performance. The limitations of PDC bits for drilling hard formations is an important technical obstacle that must be overcome before using the PDC bit to develop competii tively priced electricity from enhanced geothermal systems, as well as deep continental gas fields. Enhanced geothermal energy is a very promising source for generating electrical energy and therefore, there is an urgent need to further enhance PDC bit per-j formance in hard formations. In this paper, the cutting efficiency of the PDC bit has been) analyzed based on the development of an analytical single PDC cutter force model. The cutting efficiency of a single PDC cutterj is defined as the ratio of the volume removed by a cutter over the force required to remove that volume of rock. The cutting I efficiency is found to be a function of the back rake angle, the depth of cut and the rock property, such as the angle of internal' friction. The highest cutting efficiency is found to occur at specific back rake angles of the cutter based on the material properties of the rock. The cutting efficiency directly relates to the internal angle of friction of the rock being cut. The results of this analysis can be integrated to study PDC bit performance. It can also provide a guideline to the application' and design of PDC bits for specific rocks.

Since the Reactor Safety Study in the early 1970's, human reliability analysis (HRA) has been evolving towards a better ability to account for the factors and conditions that can lead humans to take unsafe actions and thereby provide better estimates of the likelihood of human error for probabilistic risk assessments (PRAs). The purpose of this paper is to provide an overview of recent reviews of operational events and advances in the behavioral sciences that have impacted the evolution of HRA methods and contributed to improvements. The paper discusses the importance of human errors in complex human-technical systems, examines why humans contribute to accidents and unsafe conditions, and discusses how lessons learned over the years have changed the perspective and approach for modeling human behavior in PRAs of complicated domains such as nuclear power plants. It is argued that it has become increasingly more important to understand and model the more cognitive aspects of human performance and to address the broader range of factors that have been shown to influence human performance in complex domains. The paper concludes by addressing the current ability of HRA to adequately predict human failure events and their likelihood.

The purpose of the Sandia National Laboratories (SNL) Advanced Simulation and Computing (ASC) Software Quality Plan is to clearly identify the practices that are the basis for continually improving the quality of ASC software products. Quality is defined in the US Department of Energy/National Nuclear Security Agency (DOE/NNSA) Quality Criteria, Revision 10 (QC-1) as 'conformance to customer requirements and expectations'. This quality plan defines the SNL ASC Program software quality engineering (SQE) practices and provides a mapping of these practices to the SNL Corporate Process Requirement (CPR) 001.3.6; 'Corporate Software Engineering Excellence'. This plan also identifies ASC management's and the software project teams responsibilities in implementing the software quality practices and in assessing progress towards achieving their software quality goals. This SNL ASC Software Quality Plan establishes the signatories commitments to improving software products by applying cost-effective SQE practices. This plan enumerates the SQE practices that comprise the development of SNL ASC's software products and explains the project teams opportunities for tailoring and implementing the practices.

The Ecological Footprint Model is a mechanism for measuring the environmental effects of operations at Sandia National Laboratories in Albuquerque, New Mexico (SNL/NM). This analysis quantifies environmental impact associated with energy use, transportation, waste, land use, and water consumption at SNL/NM for fiscal year 2005 (FY05). Since SNL/NM’s total ecological footprint (96,434 gha) is greater than the waste absorption capacity of its landholdings (338 gha), it created an ecological deficit of 96,096 gha. This deficit is equal to 886,470lha, or about 3,423 square miles of Pinyon-Juniper woodlands and desert grassland. 89% of the ecological footprint can be attributed to energy use, indicating that in order to mitigate environmental impact, efforts should be focused on energy efficiency, energy reduction, and the incorporation of additional renewable energy alternatives at SNL/NM.

The Arrhenius parameters for graphite oxidation in air are reviewed and compared. One-dimensional models of graphite oxidation coupled with mass transfer of oxidant are presented in dimensionless form for rectangular and spherical geometries. A single dimensionless group is shown to encapsulate the coupled phenomena, and is used to determine the effective reaction rate when mass transfer can impede the oxidation process. For integer reaction order kinetics, analytical expressions are presented for the effective reaction rate. For noninteger reaction orders, a numerical solution is developed and compared to data for oxidation of a graphite sphere in air. Very good agreement is obtained with the data without any adjustable parameters. An analytical model for surface burn-off is also presented, and results from the model are within an order of magnitude of the measurements of burn-off in air and in steam.

This report describes the successful efforts of Beacon Power to design and develop a 20-MW frequency regulation power plant based solely on flywheels. Beacon's Smart Matrix (Flywheel) Systems regulation power plant, unlike coal or natural gas generators, will not burn fossil fuel or directly produce particulates or other air emissions and will have the ability to ramp up or down in a matter of seconds. The report describes how data from the scaled Beacon system, deployed in California and New York, proved that the flywheel-based systems provided faster responding regulation services in terms of cost-performance and environmental impact. Included in the report is a description of Beacon's design package for a generic, multi-MW flywheel-based regulation power plant that allows accurate bids from a design/build contractor and Beacon's recommendations for site requirements that would ensure the fastest possible construction. The paper concludes with a statement about Beacon's plans for a lower cost, modular-style substation based on the 20-MW design.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

This report documents a series of models for describing intended and unintended discharges from liquid hydrogen storage systems. Typically these systems store hydrogen in the saturated state at approximately five to ten atmospheres. Some of models discussed here are equilibrium-based models that make use of the NIST thermodynamic models to specify the states of multiphase hydrogen and air-hydrogen mixtures. Two types of discharges are considered: slow leaks where hydrogen enters the ambient at atmospheric pressure and fast leaks where the hydrogen flow is usually choked and expands into the ambient through an underexpanded jet. In order to avoid the complexities of supersonic flow, a single Mach disk model is proposed for fast leaks that are choked. The velocity and state of hydrogen downstream of the Mach disk leads to a more tractable subsonic boundary condition. However, the hydrogen temperature exiting all leaks (fast or slow, from saturated liquid or saturated vapor) is approximately 20.4 K. At these temperatures, any entrained air would likely condense or even freeze leading to an air-hydrogen mixture that cannot be characterized by the REFPROP subroutines. For this reason a plug flow entrainment model is proposed to treat a short zone of initial entrainment and heating. The model predicts the quantity of entrained air required to bring the air-hydrogen mixture to a temperature of approximately 65 K at one atmosphere. At this temperature the mixture can be treated as a mixture of ideal gases and is much more amenable to modeling with Gaussian entrainment models and CFD codes. A Gaussian entrainment model is formulated to predict the trajectory and properties of a cold hydrogen jet leaking into ambient air. The model shows that similarity between two jets depends on the densimetric Froude number, density ratio and initial hydrogen concentration.