Queue length‐based load balancing in data center networks

Traffic load balancing in data centers is an important requirement. Traffic dynamics and possibilities of changes in the topology (e.g., failures and asymmetries) make load balancing a challenging task. Existing end‐host–based schemes either employ the predominantly used ECN or combine it with RTT to get congestion information of paths. Both congestion signals, ECN and RTT, have limitations; ECN only tells whether the queue length is above or below a threshold value but does not inform about the extent of congestion; similarly, RTT in data center networks is on the scale of up to few hundreds of microseconds, and current data center operating systems lack fine‐grained microsecond‐level timers. Therefore, there is a need of a new congestion signal which should give accurate information of congestion along the path. Furthermore, in end‐host–based schemes, detecting asymmetries in the topology is challenging due to the inability to accurately measure RTT on the scale of microseconds. This paper presents QLLB, an end‐host–based, queue length–based load balancing scheme. QLLB employs a new queue length–based congestion signal that gives an exact measure of congestion along the paths. Furthermore, QLLB uses relative‐RTT to detect asymmetries in the topology. QLLB is implemented in ns‐3 and compared with ECMP, CONGA, and Hermes. The results show that QLLB significantly improves performance of short flows over the other schemes and performs within acceptable level, of CONGA and Hermes, for long flows. In addition, QLLB effectively detects asymmetric paths and performs better than Hermes under high loads.     

A Distributed and Elastic Application Processing Model for Mobile Cloud Computing

The latest developments in mobile computing technology have increased the computing capabilities of smart mobile devices (SMDs). However, SMDs are still constrained by low bandwidth, processing potential, storage capacity, and battery lifetime. To overcome these problems, the rich resources and powerful computational cloud is tapped for enabling intensive applications on SMDs. In Mobile Cloud Computing (MCC), application processing services of computational clouds are leveraged for alleviating resource limitations in SMDs. The particular deficiency of distributed architecture and runtime partitioning of the elastic mobile application are the challenging aspects of current offloading models. To address these issues of traditional models for computational offloading in MCC, this paper proposes a novel distributed and elastic applications processing (DEAP) model for intensive applications in MCC. We present an analytical model to evaluate the proposed DEAP model, and test a prototype application in the real MCC environment to demonstrate the usefulness of DEAP model. Computational offloading using the DEAP model minimizes resources utilization on SMD in the distributed processing of intensive mobile applications. Evaluation indicates a reduction of 74.6% in the overhead of runtime application partitioning and a 66.6% reduction in the CPU utilization for the execution of the application on SMD.

     

A computationally efficient selected mapping technique for reducing PAPR of OFDM

In orthogonal frequency division multiplexing (OFDM) system, high value of peak-to-average power ratio (PAPR) is an operational problem that may cause non-linear distortion resulting in high bit error rate. Selected mapping (SLM) is a well known technique that shows good PAPR reduction capability but inflicts added computational overhead. In this paper, using Riemann sequence based SLM method, we applied reverse searching technique to find out low PAPR yielding phase sequences with significant reduction in computational complexity. Additionally, we explored side-information free transmission that achieves higher throughput but sacrifices PAPR reduction. Finally, to overcome this loss in PAPR reduction, we proposed application of Square-rooting companding technique over the output OFDM transmitted signal. Simulation results show that the proposed method is able to compensate the sacrifice in PAPR and achieved PAPR reduction of 8.9 dB with very low computational overhead.     

Piezoelectric Sensors Operating at Very High Temperatures and in Extreme Environments Made of Flexible Ultrawide-Bandgap Single-Crystalline AlN Thin Films

Extreme environments are often faced in energy, transportation, aerospace, and defense applications and pose a technical challenge in sensing. Piezoelectric sensor based on single-crystalline AlN transducers is developed to address this challenge. The pressure sensor shows high sensitivities of 0.4–0.5 mV per psi up to 900 °C and output voltages from 73.3 to 143.2 mV for input gas pressure range of 50 to 200 psi at 800 °C. The sensitivity and output voltage also exhibit the dependence on temperature due to two origins. A decrease in elastic modulus (Young's modulus) of the diaphragm slightly enhances the sensitivity and the generation of free carriers degrades the voltage output beyond 800 °C, which also matches with theoretical estimation. The performance characteristics of the sensor are also compared with polycrystalline AlN and single-crystalline GaN thin films to investigate the importance of single crystallinity on the piezoelectric effect and bandgap energy-related free carrier generation in piezoelectric devices for high-temperature operation. The operation of the sensor at 900 °C is amongst the highest for pressure sensors and the inherent properties of AlN including chemical and thermal stability and radiation resistance indicate this approach offers a new solution for sensing in extreme environments.     

Semi-Supervised Deep Learning for Multi-Tissue Segmentation from Multi-Contrast MRI

Journal of Signal Processing Systems - Segmentation of thigh tissues (muscle, fat, inter-muscular adipose tissue (IMAT), bone, and bone marrow) from magnetic resonance imaging (MRI) scans is useful...     

Scientific Workflow Makespan Minimization in Edge Multiple Service Providers Environment

The edge computing model offers an ultimate platform to support scientific and real-time workflow-based applications over the edge of the network. However, scientific workflow scheduling and execution still facing challenges such as response time management and latency time. This leads to deal with the acquisition delay of servers, deployed at the edge of a network and reduces the overall completion time of workflow. Previous studies show that existing scheduling methods consider the static performance of the server and ignore the impact of resource acquisition delay when scheduling workflow tasks. Our proposed method presented a meta-heuristic algorithm to schedule the scientific workflow and minimize the overall completion time by properly managing the acquisition and transmission delays. We carry out extensive experiments and evaluations based on commercial clouds and various scientific workflow templates. The proposed method has approximately 7.7% better performance than the baseline algorithms, particularly in overall deadline constraint that gives a success rate.

     

Kinetic modeling of pyrolysis of three Iranian waste oils in a micro-fluidized bed

Different approaches have been used to convert the waste materials into a clean syngas or other chemicals such as methanol. Among them, pyrolysis is a good candidate to produce the synthesis gas and volatile matters for industrial and refinery applications. In this work, we studied the kinetic and chemical behavior of three Iranian waste oils through a kinetic model and an experimental study. The experiments carried out in a micro-FB reactor, which is a good option for low emissions. Results showed that the reaction temperature and reaction rate are two of the most important factors for maximum conversion level of fuel. Results also showed an optimum value for reaction rate. The modeling results validated against the experimental measurements and found to be in good agreements.     

Electrochemical degradation of petroleum hydrocarbons (PAHS) from synthetic aqueous solutions

Water separated from crude oil and wastewater discharge from petroleum oil refineries contains significant quantity of dissolved hydrocarbons. Polycyclic aromatic hydrocarbons (PAHs) are major toxicants in wastewater of refineries. It is difficult to treat wastewater containing PAHs due to their recalcitrant property and low solubility. Conventional techniques for the treatment of wastewater are still a concern of toxicity. Electrochemical oxidation process has been found to be a favorable for treating wastewater. Electrodes with high stability and electrocatalytic activity are important factors for a successful electrochemical oxidation of toxic organics in wastewater. In this study titanium anodes were coated with tin, antimony and iridium oxide mixture from their respective salts by thermal decomposition method. FESEM and XRD used for surface characterization of Ti/SnO₂–Sb₂O₅–IrO₂ anode. Quantification of PAHs was done using GC–MS. Results confirm the presence of respective oxides on anode surface. Their electrocatalytic capability was tested for degradation of 16 priority PAHs in aqueous solution. Results reveal the complete degradation of naphthalene, acenaphthylene, acenaphthene and fluorene without using NaCl electrolyte. While in the presence of NaCl naphthalene, acenaphthylene, acenaphthene, fluorene, phenanthrene, anthracene, fluoranthene and pyrene were completely removed. About 98% of total PAHs removal was found at all initial pH values 3, 6, and 9 in the presence of electrolyte. Current study will be helpful in improving quality of petroleum industry wastewater containing PAHs.     

Harmonic and intermodulation performance in an R-LED series network

Analytical expressions are obtained for predicting the harmonic and intermodulation performance of R-LED series networks. These expressions are in terms of the ordinary Bessel functions with arguments depenedent on the modulation index.     

Wild melon: a novel non-edible feedstock for bioenergy

In the present research work, a non-edible oil source Cucumis melo var. agrestis(wild melon) was systematically identified and studied for biodiesel production and its characterization. The extracted oil was 29.1% of total dry seed weight. The free fatty acid value of the oil was found to be 0.64%, and the single-step alkaline transesterification method was used for conversion of fatty acids into their respective methyl esters. The maximum conversion efficiency of fatty acids was obtained at 0.4 wt% Na OH(used as catalyst), 30%(methanol to oil, v/v) methanol amount, 60 ℃ reaction temperature,600-rpm agitation rate and 60-min reaction time. Under these optimal conditions, the conversion efficiency of fatty acid was 92%. However, in the case of KOH as catalyst, the highest conversion(85%) of fatty acids was obtained at 40%methanol to oil ratio, 1.28 wt% KOH, 60 ℃ reaction temperature, 600-rpm agitation rate and 45 min of reaction time.Qualitatively, biodiesel was characterized through Fourier transform infrared spectroscopy(FTIR) and gas chromatography and mass spectroscopy(GC–MS). FTIR results demonstrated a strong peak at 1742 cm~(-1), showing carbonyl groups(C=O)of methyl esters. However, GC–MS results showed the presence of twelve methyl esters comprised of lauric acid, myristic acid, palmitic acid, non-decanoic acid, hexadecanoic acid, octadecadienoic acid and octadecynoic acid. The fuel properties were found to fall within the range recommended by the international biodiesel standard, i.e., American Society of Testing Materials(ASTM): flash point of 91 ℃, density of 0.873 kg/L, viscosity of 5.35 c St, pour point of-13 ℃, cloud point of-10 ℃, total acid number of 0.242 mg KOH/g and sulfur content of 0.0043 wt%. The present work concluded the potential of wild melon seed oil as excellent non-edible source of bioenergy.