A high‐performance sorting algorithm for multicore single‐instruction multiple‐data processors

Many sorting algorithms have been studied in the past, but there are only a few algorithms that can effectively exploit both single‐instruction multiple‐data (SIMD) instructions and thread‐level parallelism. In this paper, we propose a new high‐performance sorting algorithm, called aligned‐access sort (AA‐sort), that exploits both the SIMD instructions and thread‐level parallelism available on today's multicore processors. Our algorithm consists of two phases, an in‐core sorting phase and an out‐of‐core merging phase. The in‐core sorting phase uses our new sorting algorithm that extends combsort to exploit SIMD instructions. The out‐of‐core algorithm is based on mergesort with our novel vectorized merging algorithm. Both phases can take advantage of SIMD instructions. The key to high performance is eliminating unaligned memory accesses that would reduce the effectiveness of SIMD instructions in both phases. We implemented and evaluated the AA‐sort on PowerPC 970MP and Cell Broadband Engine platforms. In summary, a sequential version of the AA‐sort using SIMD instructions outperformed IBM's optimized sequential sorting library by 1.8 times and bitonic mergesort using SIMD instructions by 3.3 times on PowerPC 970MP when sorting 32 million random 32‐bit integers. Also, a parallel version of AA‐sort demonstrated better scalability with increasing numbers of cores than a parallel version of bitonic mergesort on both platforms. Copyright © 2011 John Wiley & Sons, Ltd.     

Stress intensity factors of an inclined elliptical crack near a bimaterial interface

In this paper the stress intensity factors are discussed for an inclined elliptical crack near a bimaterial interface. The solution utilizes the body force method and requires Green’s functions for perfectly bonded semi-infinite bodies. The formulation leads to a system of hypersingular integral equation whose unknowns are three modes of crack opening displacements. In the numerical calculation, unknown body force densities are approximated by using fundamental density functions and polynomials. The results show that the present method yields smooth variations of stress intensity factors along the crack front accurately. Distributions of stress intensity factors are presented in tables and figures with varying the shape of crack, distance from the interface, and elastic modulus ratio. It is found that the inclined crack can be evaluated by the models of vertical and parallel cracks within the error of 24% even for the cracks very close to the interface.     

Alteration of internal stresses in SiO2/Cu/TiN thin films by X-ray and synchrotron radiation due to heat treatment

A break of wiring by stress-migration becomes a problem with an integrated circuit such as LSI. The present study investigates residual stress in SiO₂/Cu/TiN film deposited on glass substrates. A TiN layer, as an undercoat, was first deposited on the substrate by arc ion plating and then Cu and SiO₂ layers were deposited by plasma coating. The crystal structure and the residual stress in the deposited multi-layer film were investigated using in-lab. X-ray equipment and a synchrotron radiation device that emits ultra-high-intensity X-rays. It was found that the SiO₂ film was amorphous and both the Cu and TiN films had a strong {1 1 1} orientation. The Cu and TiN layers in the multi thick (Cu and TiN:1.0 μm)-layer film and multi thin (0.1 μm)-layer film exhibited tensile residual stresses. Both tensile residual stresses in the multi thin-layer film are larger than the multi thick-layer film. After annealing at 400 °C, these tensile residual stresses in both the films increased with increasing the annealing temperature. Surface swelling formations, such as bubbles were observed in the multi thick-layer film. However, in the case of the multi thin-layer films, there was no change in the surface morphology following heat-treatment.     

Inside Front Cover: Fabrication of Ideally Ordered Nanoporous Alumina Films and Integrated Alumina Nanotubule Arrays by High‐Field Anodization (Adv. Mater. 17/2005)

A simple, low‐cost approach to fabricating large‐area highly ordered nanoporous alumina films in sulfuric acid solutions through a single‐step high‐field anodization, without the assistance of any additional process, is reported on p. 2115 by Chu and co‐workers. The critical high anodizing potential in the adopted electrolyte system increases with the ageing of solutions after a long period of anodization. Correspondingly, the applicable current density for stable anodization rises significantly, thus leading to high‐speed film growth. Uniform porous anodic alumina films in sulfuric acid solutions under a high electric field of 40–70 V and 1600–2000 A m^–2 are achieved.     

Shape from Shading with Interreflections Under a Proximal Light Source: Distortion-Free Copying of an Unfolded Book

We address the problem of recovering the 3D shape of an unfolded book surface from the shading information in a scanner image. This shape-from-shading problem in a real world environment is made difficult by a proximal, moving light source, interreflections, specular reflections, and a nonuniform albedo distribution. Taking all these factors into account, we formulate the problem as an iterative, non-linear optimization problem. Piecewise polynomial models of the 3D shape and albedo distribution are introduced to efficiently and stably compute the shape in practice. Finally, we propose a method to restore the distorted scanner image based on the reconstructed 3D shape. The image restoration experiments for real book surfaces demonstrate that much of the geometric and photometric distortions are removed by our method.     

1.3-μm InP-InGaAsP lasers fabricated on Si substrates by waferbonding

1.3-μm InP-InGaAsP lasers have been successfully fabricated on Si substrates by wafer bonding. InP-InGaAsP thin epitaxial films are prepared by selective etching of InP substrates and then bonded to Si wafers, after which the laser structures are fabricated on the bonded thin films. The bonding temperature has been optimized to be 400°C by considering bonding strength, quality of the bonded crystal, and compatibility with device processes. Room-temperature continuous-wave (RT CW) operation has been achieved for 6-μm-wide mesa lasers with a threshold current of 39 mA, which is identical to that of conventional lasers on InP substrates. Additionally, the lasers fabricated on Si have exhibited higher output powers than the lasers on InP, which is due to higher thermal conductivity of Si substrates. From these results, the wafer bonding is thought to be a promising technique to integrate optical devices on Si and implement optical interconnections between Si LSI chips     

Mass recovery adsorption refrigeration cycle—improving cooling capacity

The study investigates the performance of two-bed, silica gel-water adsorption refrigeration cycle with mass recovery process. The cycle with mass recovery can be driven by the relatively low temperature heat source. In an adsorption refrigeration cycle, the pressures in adsorber and desorber are different. The chiller with mass recovery process utilizes the pressure difference to enhance the refrigerant mass circulation. Cooling capacity and coefficient of performance (COP) were calculated by cycle simulation computer program to analyze the influences of operating conditions. The mass recovery cycle was compared with conventional cycle such as the single stage adsorption cycle in terms of cooling capacity and COP. The results show that the cooling capacity of mass recovery cycle is superior to that of conventional cycle and the mass recovery process is more effective for low regenerating temperature.