Engineering Papers⌕ Search

Engineering topics

Biegel, Bryan

Publications and source records attributed to Biegel, Bryan.

At least 73 records · Page 4

The Automatic Parallelisation of Scientific Application Codes Using a Computer Aided Parallelisation Toolkit

The shared-memory programming model is a very effective way to achieve parallelism on shared memory parallel computers. Historically, the lack of a programming standard for using directives and the rather limited performance due to scalability have affected the take-up of this programming model approach. Significant progress has been made in hardware and software technologies, as a result the performance of parallel programs with compiler directives has also made improvements. The introduction of an industrial standard for shared-memory programming with directives, OpenMP, has also addressed the issue of portability. In this study, we have extended the computer aided parallelization toolkit (developed at the University of Greenwich), to automatically generate OpenMP based parallel programs with nominal user assistance. We outline the way in which loop types are categorized and how efficient OpenMP directives can be defined and placed using the in-depth interprocedural analysis that is carried out by the toolkit. We also discuss the application of the toolkit on the NAS Parallel Benchmarks and a number of real-world application codes. This work not only demonstrates the great potential of using the toolkit to quickly parallelize serial programs but also the good performance achievable on up to 300 processors for hybrid message passing and directive-based parallelizations.

Ierotheou, C.↗

Modeling Ballistic Current Flow in Carbon Nanotube Wires

Experiments have shown carbon nanotubes (CNT) to be almost perfect conductors at small applied biases. The features of the CNT band structure, large velocity of the crossing subbands and the small number of modes that an electron close to the band center / Fermi energy can scatter into, are the reasons for the near perfect small bias conductance. We show that the CNT band structure does not help at large applied biases - electrons injected into the non crossing subbands can either be Bragg reflected or undergo Zener-type tunneling. This limits the current carrying capacity of CNT. We point out that the current carrying capacity of semiconductor quantum wires in the ballistic limit is different, owing to its band structure. The second aspect addressed is the relationship of nanotube chirality in determining the physics of metal-nanotube coupling. We show that a metallic-zigzag nanotube couples better than an armchair nanotube to a metal contact. This arises because in the case of armchair nanotubes, while the pi band couples well, the pi* band does not couple well to the metal. In the case of zigzag nanotube both crossing modes couple reasonably well to the metal. Many factors such as the role of curvature, strain and defects will play a role in determining the suitability of nanotubes as nanowires. From the limited view point of metal-nanotube coupling, we feel that metallic-zigzag nanotubes are preferable to armchair nanotubes.

Anantram, M. P.↗

A Simple Case Study of a Grid Performance System

This document presents a simple case study of a Grid performance system based on the Grid Monitoring Architecture (GMA) being developed by the Grid Forum Performance Working Group. It describes how the various system components would interact for a very basic monitoring scenario, and is intended to introduce people to the terminology and concepts presented in greater detail in other Working Group documents. We believe that by focusing on the simple case first, working group members can familiarize themselves with terminology and concepts, and productively join in the ongoing discussions of the group. In addition, prototype implementations of this basic scenario can be built to explore the feasibility of the proposed architecture and to expose possible shortcomings. Once the simple case is understood and agreed upon, complexities can be added incrementally as warranted by cases not addressed in the most basic implementation described here. Following the basic performance monitoring scenario discussion, unresolved issues are introduced for future discussion.

Aydt, Ruth↗

A Comparison of Parameter Study Creation and Job Submission Tools

We consider the differences between the available general purpose parameter study and job submission tools. These tools necessarily share many features, but frequently with differences in the way they are designed and implemented For this class of features, we will only briefly outline the essential differences. However we will focus on the unique features which distinguish the ILab parameter study and job submission tool from other packages, and which make the ILab tool easier and more suitable for use in our research and engineering environment.

DeVivo, Adrian↗

Some Improvements on Signed Window Algorithms for Scalar Multiplications in Elliptic Curve Cryptosystems

Scalar multiplication is an essential operation in elliptic curve cryptosystems because its implementation determines the speed and the memory storage requirements. This paper discusses some improvements on two popular signed window algorithms for implementing scalar multiplications of an elliptic curve point - Morain-Olivos's algorithm and Koyarna-Tsuruoka's algorithm.

Vo, San C.↗

Tools to Rapidly and Effectively Parallelise Serial FORTRAN Codes Using Either MPI or OpenMP

Significant progress has been made in hardware and software technologies, as a result the use of efficient standard message passing communication calls in parallel application codes has greatly assisted in reducing execution times. In addition, the introduction of OpenMP, an industrial standard for shared memory programming with directives addresses the issue of portability and also provides a means to deliver respectable performance. In general however, the manual parallelisation of a user's application code using either message passing or directives is still a very costly and error-prone process.

Jin, Hao-Qiang↗

Coupling of Armchair and Zigzag Tubes to a Free Electron Metal

The effect of nanotube chirality is of prime importance in determining its electronic properties. We address the issue of how chirality affects the coupling of a nanotube to metal contacts. We model coupling of armchair and zigzag nanotubes to metal contacts, in both the side- and end-contacted geometries. In the side-contacted geometry, the coupling of armchair and metallic-zigzag nanotubes to a free electron metal are significantly different. Namely, it is possible to drive a larger current through a metallic-zigzag nanotube. The predicted difference holds good when both (a) the entire circumference and (b) only a finite sector of the nanotube makes contact to the metal electrode. It might be possible to observe the predicted difference between armchair and zigzag nanotubes using gold contacts.

Anantram, M. P.↗

The State of NASA's Information Power Grid

This viewgraph presentation transfers the concept of the power grid to information sharing in the NASA community. An information grid of this sort would be characterized as comprising tools, middleware, and services for the facilitation of interoperability, distribution of new technologies, human collaboration, and data management. While a grid would increase the ability of information sharing, it would not necessitate it. The onus of utilizing the grid would rest with the users.

Johnston, William E.↗

NAS Grid Benchmarks: A Tool for Grid Space Exploration

We present an approach for benchmarking services provided by computational Grids. It is based on the NAS Parallel Benchmarks (NPB) and is called NAS Grid Benchmark (NGB) in this paper. We present NGB as a data flow graph encapsulating an instance of an NPB code in each graph node, which communicates with other nodes by sending/receiving initialization data. These nodes may be mapped to the same or different Grid machines. Like NPB, NGB will specify several different classes (problem sizes). NGB also specifies the generic Grid services sufficient for running the bench-mark. The implementor has the freedom to choose any specific Grid environment. However, we describe a reference implementation in Java, and present some scenarios for using NGB.

Frumkin, Michael↗

2D Quantum Transport Modeling in Nanoscale MOSFETs

With the onset of quantum confinement in the inversion layer in nanoscale MOSFETs, behavior of the resonant level inevitably determines all device characteristics. While most classical device simulators take quantization into account in some simplified manner, the important details of electrostatics are missing. Our work addresses this shortcoming and provides: (a) a framework to quantitatively explore device physics issues such as the source-drain and gate leakage currents, DIBL, and threshold voltage shift due to quantization, and b) a means of benchmarking quantum corrections to semiclassical models (such as density- gradient and quantum-corrected MEDICI). We have developed physical approximations and computer code capable of realistically simulating 2-D nanoscale transistors, using the non-equilibrium Green's function (NEGF) method. This is the most accurate full quantum model yet applied to 2-D device simulation. Open boundary conditions, oxide tunneling and phase-breaking scattering are treated on equal footing. Electrons in the ellipsoids of the conduction band are treated within the anisotropic effective mass approximation. Quantum simulations are focused on MIT 25, 50 and 90 nm "well- tempered" MOSFETs and compared to classical and quantum corrected models. The important feature of quantum model is smaller slope of Id-Vg curve and consequently higher threshold voltage. These results are quantitatively consistent with I D Schroedinger-Poisson calculations. The effect of gate length on gate-oxide leakage and sub-threshold current has been studied. The shorter gate length device has an order of magnitude smaller current at zero gate bias than the longer gate length device without a significant trade-off in on-current. This should be a device design consideration.

Svizhenko, Alexei↗

Quantum Mechanical Study of Nanoscale MOSFET

The steady state characteristics of MOSFETS that are of practical Interest are the drive current, off-current, dope of drain current versus drain voltage, and threshold voltage. In this section, we show that quantum mechanical simulations yield significantly different results from drift-diffusion based methods. These differences arise because of the following quantum mechanical features: (I) polysilicon gate depletion in a manner opposite to the classical case (II) dependence of the resonant levels in the channel on the gate voltage, (III) tunneling of charge across the gate oxide and from source to drain, (IV) quasi-ballistic flow of electrons. Conclusions dI/dV versus V does not increase in a manner commensurate with the increase in number of subbands. - The increase in dI/dV with bias is much smaller then the increase in the number of subbands - a consequence of bragg reflection. Our calculations show an increase in transmission with length of contact, as seen in experiments. It is desirable for molecular electronics applications to have a small contact area, yet large coupling. In this case, the circumferential dependence of the nanotube wave function dictates: - Transmission in armchair tubes saturates around unity - Transmission in zigzag tubes saturates at two.

Svizhenko, Alexei↗

Computational Nanotechnology of Molecular Materials, Electronics, and Actuators with Carbon Nanotubes and Fullerenes

The role of computational nanotechnology in developing next generation of multifunctional materials, molecular scale electronic and computing devices, sensors, actuators, and machines is described through a brief review of enabling computational techniques and few recent examples derived from computer simulations of carbon nanotube based molecular nanotechnology.

Srivastava, Deepak↗

An XML-Based Protocol for Distributed Event Services

A recent trend in distributed computing is the construction of high-performance distributed systems called computational grids. One difficulty we have encountered is that there is no standard format for the representation of performance information and no standard protocol for transmitting this information. This limits the types of performance analysis that can be undertaken in complex distributed systems. To address this problem, we present an XML-based protocol for transmitting performance events in distributed systems and evaluate the performance of this protocol.

Smith, Warren↗

Transport Modeling for Metallic Electrode: Semiconducting Nanotube Systems

Recently, current-voltage (I-V) characteristics have been reported by Collins et al. for a system with a scanning tunneling microscope (STM) tip and a carbon nanotube. The STM tip was driven forward into a film of many entangled nanotubes on a substrate, and then was retracted, so that one of nanotubes bridged the STM and the film. I-V characteristics had two different patterns for different heights. One showed large dI/ dV with V greater than 0, small dI/dV with V less than 0, and I = 0 near V = 0 (type-I), while the other showed rectification, i.e., I does not equal 0 only with V less than 0 (type-II), with the tip grounded. We propose a physical mechanism to explain the observed I-V patterns. We consider that the observed characteristics strongly reflected the nature of the tip (metal) - nanotube (semiconductor) contact. The other end of the nanotube was entangled well in the film, and simply provided a good Ohmic contact. We will argue that there are two different contact modes: vacuum gap and touching modes, depending on the presence or absence of a tiny vacuum gap d approx. 0.1 - 0.2 nm at the junction. These modes may be related to physisorption and chemisorption, respectively. Once admitting their existence, it is naturally shown that I-V characteristics are type-I in the vacuum gap mode, and type-II in the touching mode. We argue that the nanotube had to be an n-type semiconductor judging from the I-V characteristics, contrary to often observed p-type in the transistor applications, where p-type is probably due to the oxidation in air or the trapped charges in the silicon dioxide. Additional information is contained in the original extended abstract.

Yamada, Toshishige↗

2-D Modeling of Nanoscale MOSFETs: Non-Equilibrium Green's Function Approach

We have developed physical approximations and computer code capable of realistically simulating 2-D nanoscale transistors, using the non-equilibrium Green's function (NEGF) method. This is the most accurate full quantum model yet applied to 2-D device simulation. Open boundary conditions and oxide tunneling are treated on an equal footing. Electrons in the ellipsoids of the conduction band are treated within the anisotropic effective mass approximation. Electron-electron interaction is treated within Hartree approximation by solving NEGF and Poisson equations self-consistently. For the calculations presented here, parallelization is performed by distributing the solution of NEGF equations to various processors, energy wise. We present simulation of the "benchmark" MIT 25nm and 90nm MOSFETs and compare our results to those from the drift-diffusion simulator and the quantum-corrected results available. In the 25nm MOSFET, the channel length is less than ten times the electron wavelength, and the electron scattering time is comparable to its transit time. Our main results are: (1) Simulated drain subthreshold current characteristics are shown, where the potential profiles are calculated self-consistently by the corresponding simulation methods. The current predicted by our quantum simulation has smaller subthreshold slope of the Vg dependence which results in higher threshold voltage. (2) When gate oxide thickness is less than 2 nm, gate oxide leakage is a primary factor which determines off-current of a MOSFET (3) Using our 2-D NEGF simulator, we found several ways to drastically decrease oxide leakage current without compromising drive current. (4) Quantum mechanically calculated electron density is much smaller than the background doping density in the poly silicon gate region near oxide interface. This creates an additional effective gate voltage. Different ways to. include this effect approximately will be discussed.

Svizhenko, Alexei↗

Modeling of Kink-Shaped Carbon-Nanotube Schottky Diode with Gate Bias Modulation

A model is proposed for the recent gate-voltage (V(sub G)) modulation experiment of a kink-shaped carbon nanotube (NT) Schottky diode. Since larger V(sub G) increases both the forward and the reverse turn-on voltages of the diode in the experiment, we show that: (1) the rectification must occur at the kink where the metallic and the semiconducting NTs meet, and not at the electrode contact, and (2) the semiconducting NT must be n-type. The turn-on voltages are derived analytically as a function of V(sub G) with the electrode contact contribution and a good agreement is obtained with the experimental data.

Yamada, Toshishige↗

Parallel Processing of Adaptive Meshes with Load Balancing

Many scientific applications involve grids that lack a uniform underlying structure. These applications are often also dynamic in nature in that the grid structure significantly changes between successive phases of execution. In parallel computing environments, mesh adaptation of unstructured grids through selective refinement/coarsening has proven to be an effective approach. However, achieving load balance while minimizing interprocessor communication and redistribution costs is a difficult problem. Traditional dynamic load balancers are mostly inadequate because they lack a global view of system loads across processors. In this paper, we propose a novel and general-purpose load balancer that utilizes symmetric broadcast networks (SBN) as the underlying communication topology, and compare its performance with a successful global load balancing environment, called PLUM, specifically created to handle adaptive unstructured applications. Our experimental results on an IBM SP2 demonstrate that the SBN-based load balancer achieves lower redistribution costs than that under PLUM by overlapping processing and data migration.

Das, Sajal K.↗