This paper investigates the problem of Partitioning Skew in MapReduce-based system. Our studies with Hadoop, a widely used MapReduce implementation, demonstrate that the presence of partitioning skew causes a huge amo...
详细信息
This paper investigates the problem of Partitioning Skew in MapReduce-based system. Our studies with Hadoop, a widely used MapReduce implementation, demonstrate that the presence of partitioning skew causes a huge amount of data transfer during the shuffle phase and leads to significant unfairness on the reduce input among different data nodes. As a result, the applications experience performance degradation due to the long data transfer during the shuffle phase along with the computation skew, particularly in reduce phase. We develop a novel algorithm named LEEN for locality-aware and fairness-aware key partitioning in MapReduce. LEEN embraces an asynchronous map and reduce scheme. All buffered intermediate keys are partitioned according to their frequencies and the fairness of the expected data distribution after the shuffle phase. We have integrated LEEN into Hadoop-0.18.0. Our experiments demonstrate that LEEN can efficiently achieve higher locality and reduce the amount of shuffled data. More importantly, LEEN guarantees fair distribution of the reduce inputs. As a result, LEEN achieves a performance improvement of up to 40% on different workloads.
All-optical wavelength conversion is investigated using four wave mixing in silicon waveguides. The influences of nonlinear losses caused by two-photon absorption (TPA) and TPA-induced free-carrier absorption on the c...
All-optical wavelength conversion is investigated using four wave mixing in silicon waveguides. The influences of nonlinear losses caused by two-photon absorption (TPA) and TPA-induced free-carrier absorption on the conversion efficiency and phase-matching condition are analyzed. The conversion bandwidth is tempted to enhance by optimizing the waveguide geometry and using the two-pump regime. And finally an angled-polarization pump is used to realize the polarization-independent wavelength conversion by eliminating the polarization discrepancy between the TE and TM modes.
Multi-domain meshing from volumetric data is of great importance in many fields like medicine, biology and geology. This paper proposes a new approach to produce a high quality mesh with separated multiple domains. A ...
详细信息
Multi-domain meshing from volumetric data is of great importance in many fields like medicine, biology and geology. This paper proposes a new approach to produce a high quality mesh with separated multiple domains. A point cloud is generated from a preliminary mesh representing the boundary between different domains from the discrete volumetric representation used as input. A higher-order level-set method is employed to produce a quality sub-mesh from this point cloud and geometric flow is used as smoothing mechanism. A new approach to detect and curate intersections within an assembly of these 2-manifold sub-meshes by utilizing the intermediate volumetric representation is developed. The separation between sub-meshes can be controlled by the user using a gap threshold parameter. The resulting high quality multi-domain mesh is free from self- and inter-domain intersections and can be further utilized in finite element and boundary element computations. The proposed pipeline has been efficiently implemented and sample meshes have been provided for visualization.
Routine clinical myocardial perfusion (MP) PET imaging involves the use of cardiac gating only. Nonetheless, respiratory motion of the heart can considerably degrade the quality of MP images and the quantitative accur...
详细信息
Routine clinical myocardial perfusion (MP) PET imaging involves the use of cardiac gating only. Nonetheless, respiratory motion of the heart can considerably degrade the quality of MP images and the quantitative accuracy of myocardial uptake estimates. We first performed a quantitative evaluation of the degrading contributions of cardiac (C) and respiratory (R) motion, as well as non-motion factors of Rb-82 positron range, photon non-collinearity, crystal scattering and penetration. For a normal human simulated phantom, we showed that the combination of all above factors resulted in -48% underestimation of myocardial activity, while corrections for all non-motion factors resulted in 21%, 36% and 41% underestimated myocardial activities in the presence of C, R and C&R motion. This means that compensation for respiratory motion must be considered as critical towards achieving overall motion compensation and/or resolution modeling. To achieve respiratory motion compensation, we used translation motion vectors to first match respiratory-only gated images to the end-expiration reference frame. Next, for each cardiac gate, a 4D EM reconstruction algorithm was applied to the R-gated data within that cardiac phase. Three techniques were compared involving reconstructions of (a) a single R-gate only, and all R-gates (b) without and (c) with respiratory motion correction (MC). Using simulated PET data, quantitative comparisons of noise vs. bias trace-off curves indicated notable improvements for the proposed 4D respiratory MC method. Using CHO analysis as applied to the task of perfusion defect detection, ROC analysis of the three methods resulted in AUC values of 0.610±0.039, 0.645±0.038 and 0.821±0.029. The CLABROC statistical test revealed that the proposed MC technique significantly outperformed the other two methods in the task of defect detection.
A new geometric computing theory was proposed. On the definition level of geometric elements, using the Cartesian coordinates ideology as reference, 2D and 3D ″geometric algebra elements″ (or ″geometric elements″ ...
详细信息
A new geometric computing theory was proposed. On the definition level of geometric elements, using the Cartesian coordinates ideology as reference, 2D and 3D ″geometric algebra elements″ (or ″geometric elements″ for short, which could construct an upper-level element in the solving sequence) were constructed by geometry algebraization methods. On the processing level of geometries, geometric problems were solved with geometry methods, by which a geometric element solving sequence could be constructed. Directional property was introduced into geometries in this theory and geometries were represented in a unified format. They help to simplify the processing of finding the geometric element solving sequence for a geometry problem. The paper also tried to theoretically find out an integrated solution for geometry ambiguity issues, and established a unified, standardized geometry computing architecture. The Leibniz's mind-to process geometric objects with geometric language-was implemented in an indirect way!
Using the multiple-parameter Monte Carlo method, the effectiveness of the dual threshold voltage technique (DTV) in low power domino logic design is analyzed. Simulation results indicate that under significant tempe...
详细信息
Using the multiple-parameter Monte Carlo method, the effectiveness of the dual threshold voltage technique (DTV) in low power domino logic design is analyzed. Simulation results indicate that under significant temperature and process fluctuations, DTV is still highly effective in reducing the total leakage and active power consumption for domino gates with speed loss. Also, regarding power and delay characteristics, different structure domino gates with DTV have different robustness against temperature and process fluctuation.
Solving the Hamilton-Jacobi-Isaacs (HJI) equation, commonly used in H_(infinity) optimal control, is often referred to as a two-player differential game where one player tries to minimize the cost function while the o...
详细信息
ISBN:
(纸本)9781424477456
Solving the Hamilton-Jacobi-Isaacs (HJI) equation, commonly used in H_(infinity) optimal control, is often referred to as a two-player differential game where one player tries to minimize the cost function while the other tries to maximize it. In this paper, the HJI equation is formulated online and forward-in-time using a novel single online approximator (SOLA)-based scheme to achieve optimal regulation and tracking control of affine nonlinear continuous-time systems. The SOLA-based adaptive approach is designed to learn the infinite horizon HJI equation, the corresponding optimal control input, and the worst case disturbance. A novel parameter tuning algorithm is derived which not only achieves the optimal cost function, control input, and the disturbance, but also ensures the system states remain bounded during the online learning. Lyapunov methods are used to show that all signals are uniformly ultimately bounded (UUB) while ensuring the approximated signals approach their optimal values with small bounded error. In the absence of OLA reconstruction errors, asymptotic convergence to the optimal signals is demonstrated, and simulation results illustrate the effectiveness of the approach.
We investigate a four-level superradiant laser whose intensity is proportional to N2 and whose linewidth scales as 1/N2. In the absence of spontaneous emissions, we derive the stationary solutions of the atomic full c...
详细信息
We investigate a four-level superradiant laser whose intensity is proportional to N2 and whose linewidth scales as 1/N2. In the absence of spontaneous emissions, we derive the stationary solutions of the atomic full cooperativity, analyze their stabilities, and find the corresponding irreducible representation based on group U(4). We also investigate the spectra of the phase and amplitude fluctuations of the fields inside and outside the cavity and find that the amplitude fluctuations of the output fields cannot be squeezed in the low-frequency limit, which is completely different from the three-level superradiant laser system. Finally, we consider the stationary solutions corresponding to the partial cooperativity due to the symmetry breaking of U(4) induced by the atomic spontaneous emissions.
This paper presents a new interface framework which has been added to Easy Java Simualtions' environment in order to improve its graphical features for 3D modeling. These new 3D capabilities provide users a set of...
详细信息
This paper presents a new interface framework which has been added to Easy Java Simualtions' environment in order to improve its graphical features for 3D modeling. These new 3D capabilities provide users a set of new view elements which can be used to develop models with a high degree of reality. In this way, Easy Java Simulations becomes a powerful tool to easily and quickly create 3D realistic simulations.
The reproduction of the movements of a ship by automated platforms, without the use of sensors providing exact data related to the numeric variables involved, is a non-trivial matter. The creation of an artificial vis...
详细信息
ISBN:
(纸本)9781424478149
The reproduction of the movements of a ship by automated platforms, without the use of sensors providing exact data related to the numeric variables involved, is a non-trivial matter. The creation of an artificial vision system that can follow the cadence of said ship, in six axes of freedom, is the goal of this research. Considering that a real time response is a requisite in this case, it was decided to adopt a Boolean artificial neural network system that could identify and follow arbitrary interest points that could define, as a group, a model of the movement of an observed vessel. This paper describes the development of a prototype based on the Boolean perceptron model WiSARD (Wilkie, Stonham and Aleksander's Recognition Device), that is being implemented in the C programming language on a desktop computer using a regular webcam as input.
暂无评论