Embedded-Based Graphics Processing Unit Cluster Platform for Multiple Sequence Alignments.

High-end graphics processing units (GPUs), such as NVIDIA Tesla/Fermi/Kepler series cards with thousands of cores per chip, are widely applied to high-performance computing fields in a decade. These desktop GPU cards should be installed in personal computers/servers with desktop CPUs, and the cost and power consumption of constructing a GPU cluster platform are very high. In recent years, NVIDIA releases an embedded board, called Jetson Tegra K1 (TK1), which contains 4 ARM Cortex-A15 CPUs and 192 Compute Unified Device Architecture cores (belong to Kepler GPUs). Jetson Tegra K1 has several advantages, such as the low cost, low power consumption, and high applicability, and it has been applied into several specific applications. In our previous work, a bioinformatics platform with a single TK1 (STK platform) was constructed, and this previous work is also used to prove that the Web and mobile services can be implemented in the STK platform with a good cost-performance ratio by comparing a STK platform with the desktop CPU and GPU. In this work, an embedded-based GPU cluster platform will be constructed with multiple TK1s (MTK platform). Complex system installation and setup are necessary procedures at first. Then, 2 job assignment modes are designed for the MTK platform to provide services for users. Finally, ClustalW v2.0.11 and ClustalWtk will be ported to the MTK platform. The experimental results showed that the speedup ratios achieved 5.5 and 4.8 times for ClustalW v2.0.11 and ClustalWtk, respectively, by comparing 6 TK1s with a single TK1. The MTK platform is proven to be useful for multiple sequence alignments.

Observation of the rose petal effect over single- and dual-scale roughness surfaces.

Rose petals exhibit superhydrophobicity with strong adhesion to pin water drops, known as the 'petal effect.' It is generally believed that the petal effect is attributed to dual-scale roughness, that is, the surface possesses both a nanostructure and a microstructure (Feng et al 2008 Langmuir 24 4114). In this study, we demonstrate that the dual-scale roughness is not a necessary condition for a surface of the petal effect. A surface of single-scale roughness, either at the nanoscale or the microscale alone, within a certain roughness region may also exhibit the petal effect. The surface roughness plays the essential role on the wetting behavior and governs the contact angle in the Wenzel or Cassie state, as well as the contact angle hysteresis. A water drop on the surface of the petal effect under the condition of the advancing and receding contact angle would fall into, respectively, the Cassie and Wenzel state, which leads to a contact angle hysteresis large enough to pin the water drop. On both single and dual textured hydrophobic surfaces, a sequence of wetting transitions: Wenzel state → petal state (sticky superhydrophobic state) → lotus state (slippery superhydrophobic state) is consistently observed by simply increasing the surface roughness.

Diffusive and quantum effects of water properties in different states of matter.

The enthalpy, entropy, and free energy of water are important physical quantities for understanding many interesting phenomena in biological systems. However, conventional approaches require different treatments to incorporate quantum and diffusive effects of water in different states of matter. In this work, we demonstrate the use of the two-phase thermodynamic (2PT) model as a unified approach to obtain the properties of water over the whole phase region of water from short (∼20 ps) classical molecular dynamics trajectories. The 2PT model provides an effective way to separate the diffusive modes (gas-like component) from the harmonic vibrational modes (solid-like component) in the vibrational density of states (DoS). Therefore, both diffusive and quantum effect can be properly accounted for water by applying suitable statistical mechanical weighting functions to the DoS components. We applied the 2PT model to systematically examine the enthalpy, entropy, and their temperature dependence of five commonly used rigid water models. The 2PT results are found to be consistent with those obtained from more sophisticated calculations. While the thermodynamic properties determined from different water models are largely similar, the phase boundary determined from the equality of free energy is very sensitive to the small inaccuracy in the values of enthalpy and absolute entropy. The enthalpy, entropy, and diffusivity of water are strongly interrelated, which challenge further improvement of rigid water model via parameter fitting. Our results show that the 2PT is an efficient method for studying the properties of water under various chemical and biological environments.

Density functional theory-based electrochemical models for the oxygen reduction reaction: comparison of modeling approaches for electric field and solvent effects.

A series of density functional theory (DFT) based electrochemical models are applied to systematically examine the effect of solvent, local electric field, and electrode potential on oxygen reduction reaction (ORR) kinetics. Specifically, the key elementary reaction steps of molecular oxygen dissociation, molecular oxygen protonation, and reduction of a hydroxyl adsorbate to water over the Pt(111) surface were considered. The local electric field has slight influence on reaction energetics at the vacuum interface. Solvent molecules stabilize surface adsorbates, assisting oxygen reduction. A collective solvation-potential coupled effect is identified by including long range solvent-solvent interactions in the DFT model. The dominant path of the ORR reaction varies with electrode potential and among the modeling approaches considered. The potential dependent reaction path determined from the solvated model qualitatively agrees with experiment ORR kinetics.

Electronic structure models of oxygen adsorption at the solvated, electrified Pt(111) interface.

The adsorption of molecular oxygen is the first step in the oxygen reduction reaction. Influences of interfacial water structure and electrode potential on oxygen adsorption to the Pt(111) surface were evaluated using density functional theory. Two approaches were used to model the electrification of the interface, an applied homogeneous electric field and the double-reference method of Filhol, Taylor, and Neurock. The free energy change for molecular oxygen replacement of water at the surface shows qualitatively different trends between the two models. The inclusion of solvation effects and direct control of the electrode potential offered by the double-reference method lead to the conclusion that O(2) replacement of water is favorable at all potentials studied, and O(2) binding becomes more favorable with increasing potential. This trend is contrary to that observed using an external electric field model to represent the electrochemical double layer, and arises due to the compounded effect of potential on water-surface, oxygen-surface, and water-molecular oxygen interactions. These results indicate that oxygen replacement of adsorbed water does not limit the overall oxygen reduction reaction rate at a proton-exchange membrane fuel cell cathode. The impacts of aspects of model construction (number of water layers, water density) and variation of electrode potential on the O(2)-Pt(111) interaction are described.

Preparation of superhydrophobic surfaces of hierarchical structure of hybrid from nanoparticles and regular pillar-like pattern.

The hierarchical structure silica surface of inlaying silica nanoparticles along a regular pillar-like pattern is fabricated by embossing silica sol-gel precursor mixed with silica nanoparticles on glass substrates with an elastomeric mold. The substrate is further modified by a self-assembled fluorosilanated monolayer to reduce its surface energy. The advancing/receding contact angle measurements are performed to demonstrate that a water droplet on these surfaces can exhibit a transition from the Wenzel state to the Cassie state due to the addition of silica nanoparticles to enhance its surface roughness.

Contact angle hysteresis on regular pillar-like hydrophobic surfaces.

A series of pillar-like patterned silicon wafers with different pillar sizes and spacing are fabricated by photolithography and further modified by a self-assembled fluorosilanated monolayer. The dynamic contact angles of water on these surfaces are carefully measured and found to be consistent with the theoretical predictions of the Cassie model and the Wenzel model. When a water drop is at the Wenzel state, its contact angle hysteresis increases along with an increase in the surface roughness. While the surface roughness is further raised beyond its transition roughness (from the Wenzel state to the Cassie state), the contact angle hysteresis (or receding contact angle) discontinuously drops (or jumps) to a lower (or higher) value. When a water drop is at the Cassie state, its contact angle hysteresis strongly depends on the solid fraction and has nothing to do with the surface roughness. Even for a superhydrophobic surface, the contact angle hysteresis may still exhibit a value as high as 41 degrees for the solid fraction of 0.563.