Elastic management of web server clusters on distributed virtual infrastructures

Cluster‐based solutions are being widely adopted for implementing flexible, scalable, low‐cost and high‐performance web server platforms. One of the main difficulties to implement these platforms is the correct dimensioning of the cluster size, so as to satisfy variable and peak demand periods. In this context, virtualization is being adopted by many organizations as a solution not only to provide service elasticity, but also to consolidate server workloads, and improve server utilization rates. A virtualized web server can be dynamically adapted to the client demands by deploying new virtual nodes when the demand increases, and powering off and consolidating virtual nodes during periods of low demand. Furthermore, the resources from the in‐house infrastructure can be complemented with a cloud provider (cloud bursting), so that peak demand periods can be satisfied by deploying cluster nodes in the external cloud, on an on‐demand basis. In this paper, we analyze the scalability of hybrid virtual infrastructures for two different distributed web server cluster implementations: a simple web cluster serving static files and a multi‐tier web server platform running the CloudStone benchmark. Copyright © 2011 John Wiley & Sons, Ltd.     

Time-dependent changes of zinc speciation in four soils contaminated with zincite or sphalerite

The long-term speciation of Zn in contaminated soils is strongly influenced by soil pH, clay, and organic matter content as well as Zn loading. In addition, the type of Zn-bearing contaminant entering the soil may influence the subsequent formation of pedogenic Zn species, but systematic studies on such effects are currently lacking. We therefore conducted a soil incubation study in which four soils, ranging from strongly acidic to calcareous, were spiked with 2000 mg/kg Zn using either ZnO (zincite) or ZnS (sphalerite) as the contamination source. The soils were incubated under aerated conditions in moist state for up to four years. The extractability and speciation of Zn were assessed after one, two, and four years using extractions with 0.01 M CaCl(2) and Zn K-edge X-ray absorption fine structure (XAFS) spectroscopy, respectively. After four years, more than 90% of the added ZnO were dissolved in all soils, with the fastest dissolution occurring in the acidic soils. Contamination with ZnO favored the formation of Zn-bearing layered double hydroxides (LDH), even in acidic soils, and to a lesser degree Zn-phyllosilicates and adsorbed Zn species. This was explained by locally elevated pH and high Zn concentrations around dissolving ZnO particles. Except for the calcareous soil, ZnS dissolved more slowly than ZnO, reaching only 26 to 75% of the added ZnS after four years. ZnS dissolved more slowly in the two acidic soils than in the near-neutral and the calcareous soil. Also, the resulting Zn speciation was markedly different between these two pairs of soils: Whereas Zn bound to hydroxy-interlayered clay minerals (HIM) and octahedrally coordinated Zn sorption complexes prevailed in the two acidic soils, Zn speciation in the neutral and the calcareous soil was dominated by Zn-LDH and tetrahedrally coordinated inner-sphere Zn complexes. Our results show that the type of Zn-bearing contaminant phase can have a significant influence on the formation of pedogenic Zn species in soils. Important factors include the rate of Zn release from the contaminant phases and effects of the contaminant phase on bulk soil properties and on local chemical conditions around weathering contaminant particles.     

Characterization of community based-P2P systems and implications for traffic localization

In this paper, we present one of the first and most extensive characterizations of closed community-based P2P systems. Such systems are organic groups of peer-to-peer (P2P) clients, which can be joined only by users belonging to a certain network (e.g., connected to a given Internet Service Provider (ISP)). A number of factors motivate the growth of these communities, such as quality of content, anonymity of transfers, and the potential for better performance that enhances user experience. Our study is conducted in two contrasting environments—a campus network and a national ISP—located in different continents. In both cases, large-scale closed communities have been found to be the predominant P2P systems in use. We shed light both on the factors motivating the growth of such communities, and present results characterizing the extensiveness of their usage, the performance achievable by the systems, and the implications of such communities for network providers. While our findings are interesting in their own right, they also offer important lessons for ongoing research that seeks to localize traffic within ISP boundaries. In particular, our results suggest that (i) in ISPs with heterogeneous access technologies, the performance benefits to users on localizing P2P traffic is largely dependent on the degree of seed-like behavior of peers behind high-bandwidth access technologies; and (ii) while localization can reduce the traffic on Internet peering links, it has the potential to cause a significant increase in traffic on internal links of providers, potentially requiring upgrades of network links.     

The generation of arbitrary order curved meshes for 3D finite element analysis

A procedure for generating curved meshes, suitable for high-order finite element analysis, is described. The strategy adopted is based upon curving a generated initial mesh with planar edges and faces by using a linear elasticity analogy. The analogy employs boundary loads that ensure that nodes representing curved boundaries lie on the true surface. Several examples, in both two and three dimensions, illustrate the performance of the proposed approach, with the quality of the generated meshes being analysed in terms of a distortion measure. The examples chosen involve geometries of particular interest to the computational fluid dynamics community, including anisotropic meshes for complex three dimensional configurations.     

Geometric T–Ω approach to solve eddy currents coupled to electric circuits

This paper describes a systematic geometric approach to solve magneto‐quasi‐static coupled field–circuit problems. The field problem analysis is based on formulating the boundary value problem with an electric vector potential and a scalar magnetic potential. The field–circuit coupling and the definition of potentials are formally examined within the framework of homology theory. Copyright © 2007 John Wiley & Sons, Ltd.     

Real-time optimal commutation for minimizing thermally induced inaccuracy in multi-motor driven stages

In this paper minimum power commutation laws of permanent-magnet synchronous linear motors are extended to multiple-motor driven systems. A system-wide view of commutation is shown to exploit the redundancy of motor coils and motor force generation that exist in over actuated motion platforms. We propose a novel commutation law, that is only realizable by a multiple-motor system view, which minimizes total power subject to the constraint that the thermal distortion at the motion platform work space, caused by the motor coil heating, is minimized. Due to the added constraint, there is no closed-form solution and the problem is shown to be non-convex. This problem solution is proven to be equally solved by the combination of an algebraic and convex problem. To realize the commutation laws, a custom embedded solver, using an interior-point method, is described to solve the convex problem in real-time and is shown to converge to a solution, in under 35 μs, within the update rate of our system running at over 9 KHz.     

Slow formation and dissolution of Zn precipitates in soil: a combined column-transport and XAFS study

Recent spectroscopic studies have demonstrated the formation of layered double hydroxides (LDH) and phyllosilicates upon sorption of Zn2+, Ni2+, and Co2+ to clay minerals and aluminum oxides at neutral to alkaline pH and at relatively high initial metal concentrations (>1 mM). The intention of the present study was to investigate whether such phases also form in soil under slightly acidic conditions and at lower metal concentrations. Columns packed with a loamy soil were percolated with aqueous solutions containing 0.1 or 0.2 mM Zn, Ni, Co, and Cd in a 10 mM CaCl2 background at pH 6.5. Metal breakthrough curves indicated a rapid initial sorption step, resulting in retarded breakthrough fronts, followed by further slow metal retention during the entire loading period of 42 days (7000 pore volumes). Total metal sorption and the contribution of slow sorption processes decreased in the order Zn > Ni > Co > Cd. Leaching the reacted soil with 10 mM CaCl2 at pH 6.5 remobilized 8% of the total retained Zn, 15% of Ni, 21% of Co, and 77% of Cd. Subsequent leaching with acidified influent (pH 3.0) remobilized most of the remaining metals. X-ray absorption fine-structure (XAFS) spectroscopy revealed that slow Zn sorption was due to the formation of a Zn-Al LDH precipitate. Although Ni, Co, and Cd concentrations were too low for XAFS analysis, their leaching patterns suggest that part of Ni and Co were also incorporated in solid phases, while most sorbed Cd was still present as exchangeable sorption complex after 42 days. A small but significant percentage of the sorbed metals (2-5%) remained in the soil, even after leaching with more than 3000 pore volumes at pH 3.0, which may suggest micropore diffusion or incorporation into more stable mineral phases.     

Structure and thermostability of PMMA in PMMA/silica nanocomposites: Effect of high‐energy ball milling and the amount of the nanofiller

In this work, the effect of the presence of silica nanoparticles in the structure, dynamics, and thermodegradation of poly(methyl methacrylate), PMMA, has been considered. A new method for preparing nanocomposites has been considered in which high‐energy ball milling, HEBM, was used to uniformly disperse nanoparticles within a polymeric matrix (PMMA). Fourier transform infrared spectroscopy, FTIR, was used to analyze the structure and dynamics of the PMMA to study the influence of the presence of silica nanoparticles and finally to discuss the thermal stability of these nanocomposites. Dynamic thermogravimetric analyses were performed to carry out the kinetics analysis about the thermostability of the modified PMMA, being the milling process and the amount of nanoparticles the variables taken into account. The milling process creates chain ends on the PMMA favoring its first step of thermal degradation. There is only an improvement in terms of thermal stability if the amount of silica nanoparticles within the PMMA exceeds a threshold higher than 2% by weight when the unfavorable effect of the milling process is compensated. POLYM. COMPOS., 31:1585–1592, 2010. © 2009 Society of Plastics Engineers     

On the extension of k−ω−SST corrections to predict flow separation on thick airfoils with leading-edge roughness

Modern wind turbines employ thick airfoils in the outer region of the blade with strong adverse pressure gradients and high sensitivity to flow separation, which can be anticipated by leading-edge roughness. However, Reynolds average Navier-Stokes simulations currently overpredict the Reynolds shear stresses near the surface, and the flow separation is not correctly predicted. Hence, these methods are not representative enough to optimize the blade design to avoid flow separation, which becomes relevant for rough blades. While several eddy-viscosity corrections in the $k-\omega -SST$ turbulence model have been previously studied to predict flow separation over smooth airfoils, the present study aims to extend their applicability to airfoils with leading-edge roughness. Two corrections, whose effect on flow physics has not been empirically quantified, are addressed. Particle image velocimetry measurements have been performed on a 30% thick airfoil to quantify the impact of these corrections. The reduction of the eddy viscosity introduced by the corrections leads to a shift of the peak location of the Reynolds shear stresses away from the surface, which, in turn, promotes flow separation and improves the prediction of the mean velocity and the pressure-coefficient distribution. Besides, the ratio between the main turbulent shear stress and turbulent kinetic energy is demonstrated to be lower than the standard value used in the $k-\omega -SST$ turbulence model at the boundary-layer outer edge. Adjusting this ratio for an angle of attack of 0° decreases the error on the predicted lift and drag coefficients from 75% to 3% and from 58% to 39%, respectively.