A self‐scalable load injection service

Load testing of applications is an important and costly activity for software provider companies. Classical solutions are very difficult to set up statically, and their cost is prohibitive in terms of both human and hardware resources. Virtualized cloud computing platforms provide new opportunities for stressing an application's scalability, by providing a large range of flexible and less expensive (pay‐per‐use model) computation units. On the basis of these advantages, load testing solutions could be provided on demand in the cloud. This paper describes a Benchmark‐as‐a‐Service solution that automatically scales the load injection platform and facilitates its setup according to load profiles. Our approach is based on: (i) virtualization of the benchmarking platform to create self‐scaling injectors; (ii) online calibration to characterize the injector's capacity and impact on the benched application; and (iii) a provisioning solution to appropriately scale the load injection platform ahead of time. We also report experiments on a benchmark illustrating the benefits of this system in terms of cost and resource reductions. Copyright © 2013 John Wiley & Sons, Ltd.     

Drift hole mobility in strained and unstrained doped Si1-xGex alloys

Results of the drift hole mobility in strained and unstrained SiGe alloys are reported for Ge fractions varying from 0 to 30% and doping levels of 10¹⁵-10¹⁹ cm^-3. The mobilities are calculated taking into account acoustic, optical, alloy, and ionized-impurity scattering. The mobilities are then compared with experimental results for a boron doping concentration of 2×10¹⁹ cm^-3. Good agreement between experimental and theoretical values is obtained. The results show an increase in the mobility relative to that of silicon     

Artificial neural networks for coordination and control: The portability of experiential representations

It is time to locate connectionist representation theory in the new wave of robotics research. The utility of representations developed in artificial neural networks (ANNs) during learning has been demonstrated in cognitive science research since the 1980s. The research reported here puts learned representations to work in a decentered control task, the disembodied arm problem, in which a mobile robot operates an arm fixed to a table to pick up objects. There is no physical linkage between the arm and the robot and so the robot's point of view must be decentered. This is done by developing a modular Artificial Neural Net system in three stages: (i) a classifier net is trained with laser scan data to output transformationally invariant position classes; (ii) an arm net is trained for picking up objects; (iii) an inter net is trained to communicate and coordinate the sensing and acting. The completed system is shown to create new nonsymbolic transformationally invariant representations in order to perform the effective generalization of decentered viewpoints.     

Crude-oil price volatility and agricultural employment in the USA

This study begins by asking whether fluctuations in the price of crude oil have affected agricultural employment in the USA. After reviewing previous assessments of the issue, the existence of an empirical relationship between agricultural employment and crude-oil price volatility is established using Granger causality. Subsequently, the nature of the relationship is estimated with the results suggesting that at least three full years are required before the measurable impacts of a percentage change in the real price of crude oil on the change in agricultural employment are exhausted. Finally, the structural stability of the functional relationship between the change in agricultural employment and the volatility of the price of crude oil, the percentage changes in expected net farm income, realized technological innovation, and the wage rate are examined.     

Effect of alkyl chain length on the electrochemical perfluorination of n-alkane (C6–C10) carboxylic acid chlorides

Electrochemical perfluorination (ECPF) of n-hexanoyl, n-heptanoyl, n-octanoyl, n-nonanoyl and n-decanoyl chlorides was carried out under identical experimental conditions in liquid HF. The product distribution among perfluorinated carboxylic acids, perfluoro ethers, perfluoroalkanes, isomerised and fragmented products containing less number of carbon atoms was identified using ¹⁹F NMR. The selectivity of C₆–C₁₀ perfluoro carboxylic acid varied between 29 and 36%. The alkali insoluble perfluoro cyclic ether and perfluoro alkane fractions increased with increasing chain length. The increase of perfluoroalkane fractions is mainly due to decarboxylation. Cyclic ether fractions also decreased slightly with increase in chain length. Among the cyclic ethers α substituted oxolanes were the predominant products. Six membered cyclic ethers were always found to contain β substitution. The possible pathways for these products are also indicated. An erratum to this article can be found at     

Matching treatment focus to patient social investment and support: 18-month follow-up results.

Patients were randomly assigned to 1 of 3 treatments: brief broad-spectrum (BBS), extended relationship enhancement (ERE), or extended cognitive–behavioral (ECB). A hierarchical latent growth model was used to analyze the data of 188 patients (82%) followed for 18 months. ERE treatment was significantly more effective in increasing abstinence of patients entering treatment with a network unsupportive of abstinence or with a low level of investment in their network, whereas BBS treatment was more effective for patients with either (a) both a social network unsupportive of abstinence and a low level of network investment or (b) high investment in a network supportive of abstinence. ECB outcomes were neither as good as those matched nor as bad as those mismatched to the different exposures of relationship enhancement. This suggests that dose of relationship enhancement should be determined after assessing patient relationships. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Electrochemical behaviour of Mn3+/Mn2+, Co3+/Co2+ and Ce4+/Ce4+ redox mediators in methanesulfonic acid

The redox behaviour of Mn³⁺/Mn²⁺, Co³⁺/Co²⁺ and Ce⁴⁺/Ce³⁺ mediators commonly used in indirect oxidation of organic compounds were evaluated in methane sulfonic acid on a glassy carbon working electrode employing cyclic voltammetry. Manganic methanesulfonate exhibits higher instability in dilute methanesulfonic acid. The solid MnO₂ formed during disproportionation on the glassy carbon electrode further affects the reproducibility. Cobaltic methanesulfonate formation occurs only at oxygen evolution region rendering the overall oxidation process less efficient. Ceric methane sulfonate formation is highly efficient over a wide acid concentration range. Ceric methanesulfonate can also be employed over a wide temperature range to oxidize different aromatic compounds.     

Evaluation of chromic oxide,glycerol triether,and β-sitosterol as fecal flow markers in two species of nonhuman primates

Recovery of β-sitosterol, glycerol triether (1-hexadecyl-2,3,-didodecyl glycerol triether), and chromic oxide was studied in African green monkeys and stumptail macaques consuming diets containing 0.75 mg/Cal cholesterol and 38% of calories as safflower oil or butter. Following oral administration of these compounds, feces were collected daily for 9 days. For all animals, excretion of β-sitosterol and glycerol triether paralleled one another almost exactly. Except for two animlas, this was also true for chromic acid. Essentially 100% of the administered β-sitosterol and 90–95% of the glycerol triether were recovered; excretion of these markers virtually was complete by day 3. Ninety-two percent of the β-sitosterol was isolated in the nonsaponifiable lipid extract of the feces with less than 6% in the remaining aqueous phase. A maximum of 3.1% of the β-sitosterol and 1.8% of the glycerol triether were found in the blood. For stumptail macaques, the major excretory form of β-sitosterol was the 5β-derivative. African green monkeys were more variable; one animal excreted the bulk of the β-sitosterol unchanged while others excreted greater than 80% as the ring-saturated 5β-derivative. Animals consuming the safflower oil containing diet consistently excreted a greater percentage of the β-sitosterol unchanged, compared with those animals eating the butter containing diet. There was no evidence for steroid ring degradation in any of the animals used in this study.     

Inheritance Of Resistance To Mammalian Herbivores and Of Plant Defensive Chemistry In A <Emphasis Type="Italic">Eucalyptus</Emphasis> Species

Hybridization in plants provides an opportunity to investigate the patterns of inheritance of hybrid resistance to herbivores, and of the plant mechanisms conferring this resistance such as plant secondary metabolites. We investigated how inter-race differences in resistance of Eucalyptus globulus to a generalist mammalian herbivore, Trichosurus vulpecula, are inherited in their F₁ hybrids. We assessed browsing damage of three-year-old trees in a common environment field trial on four hybrid types of known progeny. The progency were artificial intra-race crosses and reciprocal inter-race F₁ hybrids of two geographically distinct populations (races) of E. globulus; north-eastern Tasmania and south-eastern Tasmania. Populations of trees from north-eastern Tasmania are relatively susceptible to browsing by T. vulpecula, while populations from south-eastern Tasmania are more resistant. We assessed the preferences of these trees in a series of paired feeding trials with captive animals to test the field trial results and also investigated the patterns of inheritance of plant secondary metabolites. Our results demonstrated that the phenotypic expression of resistance of the inter-race F₁ hybrids supported the additive pattern of inheritance, as these hybrids were intermediate in resistance compared to the pure parental hybrids. The expression of plant secondary metabolites in the F₁ hybrids varied among groups of individual compounds. The most common pattern supported was dominance towards one of the parental types. Together, condensed tannins and essential oils appeared to explain the observed patterns of resistance among the four hybrid types. While both chemical groups were inherited in a dominant manner in the inter-race F₁ hybrids, the direction of dominance was opposite. Their combined concentration, however, was inherited in an additive manner, consistent with the phenotypic differences in browsing.     

Towards adaptive learning with improved convergence of deep belief networks on graphics processing units

In this paper we focus on two complementary approaches to significantly decrease pre-training time of a deep belief network (DBN). First, we propose an adaptive step size technique to enhance the convergence of the contrastive divergence (CD) algorithm, thereby reducing the number of epochs to train the restricted Boltzmann machine (RBM) that supports the DBN infrastructure. Second, we present a highly scalable graphics processing unit (GPU) parallel implementation of the CD-k algorithm, which boosts notably the training speed. Additionally, extensive experiments are conducted on the MNIST and the HHreco databases. The results suggest that the maximum useful depth of a DBN is related to the number and quality of the training samples. Moreover, it was found that the lower-level layer plays a fundamental role for building successful DBN models. Furthermore, the results contradict the pre-conceived idea that all the layers should be pre-trained. Finally, it is shown that by incorporating multiple back-propagation (MBP) layers, the DBNs generalization capability is remarkably improved.