Geometric Generalizations of the Power of Two Choices

A well-known paradigm for load balancing in distributed systems is the``power of two choices,''whereby an item is stored at the less loaded of two (or more) random alternative servers. We investigate the power of two choices in natural settings for distributed computing where items and servers reside in a geometric space and each item is associated with the server that is its nearest neighbor. This is in fact the backdrop for distributed hash tables such as Chord, where the geometric space is determined by clockwise distance on a one-dimensional ring. Theoretically, we consider the following load balancing problem. Suppose that servers are initially hashed uniformly at random to points in the space. Sequentially, each item then considers d candidate insertion points also chosen uniformly at random from the space,and selects the insertion point whose associated server has the least load. For the one-dimensional ring, and for Euclidean distance on the two-dimensional torus, we demonstrate that when n data items are hashed to n servers,the maximum load at any server is log log n / log d + O(1) with high probability. While our results match the well-known bounds in the standard setting in which each server is selected equiprobably, our applications do not have this feature, since the sizes of the nearest-neighbor regions around servers are non-uniform. Therefore, the novelty in our methods lies in developing appropriate tail bounds on the distribution of nearest-neighbor region sizes and in adapting previous arguments to this more general setting. In addition, we provide simulation results demonstrating the load balance that results as the system size scales into the millions.

Testing the resolving power of two-dimensional K+K+ interferometry

We show results from an analysis performed to test the resolving power of a two-dimensional χ2 method proposed previously when applied to the case of kaon interferometry, where no significant contribution from long-lived resonances is expected. For that purpose, use is made of the preliminary E859 K+K+ interferometry data from Si+Au collisions at 14.6/4 GeV/c. Although less sensitivity is achieved in the present case, this analysis seems to favor scenarios with no resonance formation at the AGS energy range. The possible compatibility of data with zero decoupling proper time interval, conjectured by the three-dimensional experimental analysis, is also investigated and is ruled out when considering more realistic dynamical models with expanding sources. Furthermore, these results strongly emphasize that the static Gaussian parametrization cannot be trusted under more realistic conditions, leading to a distorted or even wrong interpretation of the source parameters.

Resolving power of 2D pion interferometry

A χ2 analysis is performed to test the resolving power of two-dimensional pion interferometry using for illustration the preliminary E802 data on Si+Au at 14.6 A GeV/c. We find that the resolving power to distinguish two decoupling geometries of different dynamical models is enhanced by studying the variation of the mean χ2 per degrees of freedom with respect to the range of the analysis in the qT, qL plane. The preliminary data seem to rule out dynamical models with significant ω, η resonance formation yields.

Algoritmos genéticos aplicados ao projeto de filtros com coeficientes em soma de potências de dois

Esta dissertação tem como objetivo aplicar um algoritmo genético (GA) ao projeto de filtros FIR com coeficientes quantizados representados em somas de potências de dois com sinal (SPT). Os filtros FIR apresentam configurações que permitem a obtenção de fase linear, atributo desejado em diversas aplicações que necessitam de atraso de grupo constante. A representação SPT, de fácil implementação em circuitos, foi discutida e uma comparação das representações SPT mínimas e canônicas foi feita, baseada no potencial de redução de operações e na variedade de valores representáveis. O GA é aplicado na otimização dos coeficientes SPTs do filtro, para que este cumpra as suas especificações de projeto. Foram feitas análises sobre o efeito que diversos parâmetros do GA como a intensidade de seleção, tamanho das populações, cruzamento, mutação, entre outros, têm no processo de otimização. Foi proposto um novo cruzamento que produz a recombinação dos coeficientes e que obteve bons resultados. Aplicou-se o algoritmo obtido na produção de filtros dos tipos passa-baixas, passa-altas, passa-faixas e rejeita-faixas.

Performance study of two different compensating devices in a custom power park

Simulation study of a custom power park (CPP) is presented. It is assumed that the park contains unbalanced and nonlinear loads in addition to a sensitive load. Two different types of compensators are used separately to protect the sensitive load against unbalance and distortion caused by the other loads. It has been shown that a shunt compensator can regulate the voltage of the CPP bus, whereas the series compensator can only regulate the sensitive load terminal voltage. Additional issues such as the load transfer through a static transfer switch, detection of sag/fault etc. are also discussed. The concepts are validated through PSCAD/EMTDC simulation studies on a sample distribution system.

Adhesive behavior of two-dimensional power-law graded materials

In this paper, we investigate the adhesive contact between a rigid cylinder of radius R and a graded elastic half-space with a Young's modulus varying with depth according to a power-law, E = E-0(y/c(0))(k) (0 < k < 1), while the Poisson's ratio v remains constant. The results show that, for a given value of ratio R/C-0, a critical value of k exists at which the pull-off force attains a maximum; for a fixed value of k, the larger the ratio R/c(0), the larger the pull-off force is. For Gibson materials (i.e., k = 1 and v = 0.5), closed-form analytical solutions can be obtained for the critical contact half-width at pull-off and pull-off force. We further discuss the perfect stick case with both externally normal and tangential loads.

Phase-locking of two large-core fibre lasers with 60W of coherent output power

A phase-locking fibre laser array with up to 60 W of coherent output power based on two large-core fibre is reported. The slope efficiency of the in-phase mode is 37%. For two cases of spacings between the cores, steady high-contrast interference stripes are observed. When the whole system operates under a high pump power level, no thermal effects for the spatial filter have been observed, which means that we can increase the coherent output power further by increasing the individual fibre laser power.

Comparison of two kinds of active layers for high-power narrow-stripe distributed feedback laser diodes

Usually GaAs/AlGaAs is utilized as an active layer material in laser diodes operating in the spectral range of 800 850 nm. In this work, in addition to a traditional unstrained GaAs/AlGaAs distributed feedback (DFB) laser diode, a compressively strained InGaAlAs/AlGaAs DFB laser diode is numerically investigated in characteristic. The simulation results show that the compressively strained DFB laser diode has a lower transparency carrier density, higher gain, lower Auger recombination rate, and higher stimulated recombination rate, which lead to better a device performance, than the traditional unstrained GaAs/AlGaAs DFB laser diode.

Contrastive study of two SESAMs for passive mode-locking in Nd:YVC>4 laser with low pump power

Two semiconductor saturable absorber mirrors (SESAMs), of which one is coated with 50% reflection film on the top and the other is not, were contrastively studied in passively mode-locked solid-state lasers which were pumped by low output power laser diode (LD). Experiments have shown that reducing the modulation depth of SESAM by coating partial reflection film, whose reflectivity is higher than that between SESAM and air interface, is an effective method to get continuous wave (CW) mode-locking instead of Q-switched mode-locking (QML) in low power pumped solid-state lasers. A simple Nd:YVO4 laser pumped by low power LD, in which no water-cooling system was used, could obtain CW mode-locking by the 50% reflector coated SESAM with average output power of ~ 20 mW

Non-modal instabilities of two-dimensional disturbances in plane Couette flow of a power-law fluid

Instabilities of fluid flows have traditionally been investigated by normal mode analysis, i.e. by linearizing the equations of flow and testing for unstable eigenvalues of the linearized problem. However, the results of eigenvalue analysis agree poorly in many cases with experiments, especially for shear flows. In this paper we study the instabilities of two-dimensional Couette flow of a polymeric fluid in the framework of non-modal stability theory rather than normal mode analysis. A power-law model is used to describe the polymeric liquid. We focus on the response to external excitations and initial conditions by examining the pseudospectra structures and the transient energy growths. For both Newtonian and non-Newtonian flows, the results show that there can be a rather large transient growth even though the linear operator of Couette flow has no unstable eigenvalue. The effects of non-Newtonian viscosity on the transient behaviors are examined in this study. The results show that the "shear-thinning/shear-thickening" effect increases/decreases the amplitude of responses to external excitations and initial conditions. (C) 2010 Elsevier B.V. All rights reserved.

The power of NMR: in two and three dimensions

Two dimensional NMR experiments use a sequence of two or more pulses with a variable time delay to generate spectra. COSY spectra clarify where the protons are in a molecule. Two and three dimensional NMR are used to solve protein structures.

The power of NMR.: in two and three dimensions

Soon after its discovery in the 1950s, NMR had become an indispensable tool fr chemists. In the 1970s and 1980s, the power of the technique was extended from one dimension to two and even three dimensions, opening up exciting applkications in both chemistry and biochemistry. the success of one dimensional. high-resolution NMR stems from the unique insights that it can provide about molecular structure. The chemical shift of a nucleus gives invaluable information abut the chemical environment in which that nucleus is located, Coupling interactions between hydorgen nuclei, as revealed by characteristic splitting patterns inthe 1H-NMR spectrum, provide informaton about the loaction of one group of hydorgen atoms relative to others inthe molecule. And the nuclearf Overhauser effect (nOe) can shed light on molecular stereochemistry.

Fields of two power conductor

The goal of this work is find a description for fields of two power conductor. By the Kronecker-Weber theorem, these amounts to find the subfields of cyclotomic field $\mathbb{Q}(\xi_{2^r})$, where $\xi_{2^r}$ is a $2^r$-th primitive root of unit and $r$ a positive integer. In this case, the cyclotomic extension isn't cyclic, however its Galois group is generated by two elements and the subfield can be expressed by $\mathbb{Q}(\theta)$ for a $\theta\in\mathbb{Q}(\xi_{2^r})$ convenient.