PLAStiCC: Predictive Look-Ahead Scheduling for Continuous dataflows on Clouds

Scalable stream processing and continuous dataflow systems are gaining traction with the rise of big data due to the need for processing high velocity data in near real time. Unlike batch processing systems such as MapReduce and workflows, static scheduling strategies fall short for continuous dataflows due to the variations in the input data rates and the need for sustained throughput. The elastic resource provisioning of cloud infrastructure is valuable to meet the changing resource needs of such continuous applications. However, multi-tenant cloud resources introduce yet another dimension of performance variability that impacts the application's throughput. In this paper we propose PLAStiCC, an adaptive scheduling algorithm that balances resource cost and application throughput using a prediction-based lookahead approach. It not only addresses variations in the input data rates but also the underlying cloud infrastructure. In addition, we also propose several simpler static scheduling heuristics that operate in the absence of accurate performance prediction model. These static and adaptive heuristics are evaluated through extensive simulations using performance traces obtained from Amazon AWS IaaS public cloud. Our results show an improvement of up to 20% in the overall profit as compared to the reactive adaptation algorithm.

Spotting 2D atomic layers on aluminum nitride thin films

Substrates for 2D materials are important for tailoring their fundamental properties and realizing device applications. Aluminum nitride (AIN) films on silicon are promising large-area substrates for such devices in view of their high surface phonon energies and reasonably large dielectric constants. In this paper epitaxial layers of AlN on 2 `' Si wafers have been investigated as a necessary first step to realize devices from exfoliated or transferred atomic layers. Significant thickness dependent contrast enhancements are both predicted and observed for monolayers of graphene and MoS2 on AlN films as compared to the conventional SiO2 films on silicon, with calculated contrast values approaching 100% for graphene on AlN as compared to 8% for SiO2 at normal incidences. Quantitative estimates of experimentally measured contrast using reflectance spectroscopy show very good agreement with calculated values. Transistors of monolayer graphene on AlN films are demonstrated, indicating the feasibility of complete device fabrication on the identified layers.

Quantitative characterization of adhesion and stiffness of corneal lens of Drosophila melanogaster using atomic force microscopy

Atomic force Microscopy (AFM) has become a versatile tool in biology due to its advantage of high-resolution imaging of biological samples close to their native condition. Apart from imaging, AFM can also measure the local mechanical properties of the surfaces. In this study, we explore the possibility of using AFM to quantify the rough eye phenotype of Drosophila melanogaster through mechanical properties. We have measured adhesion force, stiffness and elastic modulus of the corneal lens using AFM. Various parameters affecting these measurements like cantilever stiffness and tip geometry are systematically studied and the measurement procedures are standardized. Results show that the mean adhesion force of the ommatidial surface varies from 36 nN to 16 nN based on the location. The mean stiffness is 483 +/- 5 N/m, and the elastic modulus is 3.4 +/- 0.05 GPa (95% confidence level) at the center of ommatidia. These properties are found to be different in corneal lens of eye expressing human mutant tau gene (mutant). The adhesion force, stiffness and elastic modulus are decreased in the mutant. We conclude that the measurement of surface and mechanical properties of D. melanogaster using AFM can be used for quantitative evaluation of `rough eye' surface. (C) 2015 Elsevier Ltd. All rights reserved.

The Uniaxial Tensile Deformation of Ni Nanowire: Atomic-Scale Computer Simulations

The mechanical deformations of nickel nanowire subjected to uniaxial tensile strain at 300 K are simulated by using molecular dynamics with the quantum corrected Sutten-Chen many-body force field. We have used common neighbor analysis method to investigate the structural evolution of Ni nanowire during the elongation process. For the strain rate of 0.1%/ps, the elastic limit is up to about 11% strain with the yield stress of 8.6 GPa. At the elastic stage, the deformation is carried mainly through the uniform elongation of the distances between the layers (perpendicular to the Z-axis) while the atomic structure remains basically unchanged. With further strain, the slips in the {111} planes start to take place in order to accommodate the applied strain to carry the deformation partially, and subsequently the neck forms. The atomic rearrangements in the neck region result in a zigzag change in the stress-strain curve; the atomic structures beyond the region, however, have no significant changes. With the strain close to the point of the breaking, we observe the formation of a one-atom thick necklace in Ni nanowire. The strain rates have no significant effect on the deformation mechanism, but have some influence on the yield stress, the elastic limit, and the fracture strain of the nanowire.

Surface atomic properties of tetrahedral amorphous carbon

The surface energy and surface atomic structure of tetrahedral amorphous carbon has been calculated by an ab-initio method. The surface atoms are found to reconstruct into sp2 sites often bonded in graphitic rings. Placing the dangling bonds on adjacent surface atoms lower their energy by π-bonding and this is the source of the low surface energy. The even lower surface energy of hydrogenated amorphous carbon (a-C:H) is due to the hydrogenation of all broken surface bonds. © 2005 Elsevier B.V. All rights reserved.

Experimental Investigation of the Velocity Effect on Adhesion Forces with an Atomic Force Microscope Atomic Force Microscope

Capillary forces are significantly dominant in adhesive forces measured with an atomic force microscope (AFM) in ambient air, which are always thought to be dependent on water film thickness, relative humidity, and the free energy of water film. We study the nature of the pull-off force on a variety of surfaces as a function of tip velocity. It is found that the capillary forces are of relatively strong dependence on tip velocity. The present experiment is expected to provide a better understanding of the work mechanism of AFM in ambient air.

Competitive protein adsorption studied with atomic force microscopy and imaging ellipsometry

The adsorption and competitive adsorption of collagen and bovine serum albumin (BSA) were directly visualized and quantified using atomic force microscopy (AFM) and imaging ellipsometry. Chemically modified silicon surfaces were used as hydrophilic and hydrophobic substrates. The results showed that collagen and BSA in single component solution adsorbed onto a hydrophobic surface two times more than that onto a hydrophilic surface. The competitive adsorption between collagen and BSA showed that serum albumin preferentially adsorbed onto a hydrophobic surface, while collagen on a hydrophilic surface. In the binary solution of BSA (1 mg/ml BSA) and collagen (0.1 mg/ml), nearly 100% of the protein adsorbed onto the hydrophobic surface was BSA, but on the hydrophilic surface only about 6% was BSA. Surface affinity was the main factor controlling the competitive adsorption.

Studying Aggregate in Lysozyme Solution by Atomic Force Microscope

The aggregates in lysozyme solution with different NaCl concentration were investigated by Atomic Force Microscope (AFM). The AFM images show that there exist lysozyme monomers, n-mers and clusters in lysozyme solution when the conditions are not suitable for crystal growth. In favorable conditions for crystal growth, the lysozyme clusters disappear and almost only monomers exist in solution.

Concentrated-Mass Cantilever Enhances Multiple Harmonics In Tapping-Mode Atomic Force Microscopy

The natural frequencies of a cantilever probe can be tuned with an attached concentrated mass to coincide with the higher harmonics generated in a tapping-mode atomic force microscopy by the nonlinear tip-sample interaction force. We provide a comprehensive map to guide the choice of the mass and the position of the attached particle in order to significantly enhance the higher harmonic signals containing information on the material properties. The first three eigenmodes can be simultaneously excited with only one carefully positioned particle of specific mass to enhance multiple harmonics. Accessing the interaction force qualitatively based on the high-sensitive harmonic signals combines the real-time material characterization with the imaging capability. (C) 2008 American Institute of Physics.