Sharif Digital Repository

Nowadays MapReduce and its open source implementation, Apache Hadoop, are the most widespread solutions for handling massive dataset on clusters of commodity hardware. At the expense of a somewhat reduced performance in comparison to HPC technologies, the MapReduce framework provides fault tolerance and automatic parallelization without any efforts by developers. Since in many cases Hadoop is adopted to support business critical activities, it is often important to predict with fair confidence the execution time of submitted jobs, for instance when SLAs are established with end-users. In this work, we propose and validate a hybrid approach exploiting both queuing networks and support vector... 

Abstract Elastic optical networking has recently been proposed for use in optical transport networks to cope with increasingly heterogeneous and dynamic demand patterns. In this paper, we study the blocking performance of a multi-class elastic optical link for which a demand needs to be allocated a contiguous subset of the entire spectrum. This problem is different than the well-known blocking problem in multi-class multi-server loss systems due to the contiguous allocation constraint. We first propose a non-work-conserving aligned spectrum allocation policy which is shown to outperform the conventional first fit-based work-conserving allocation policy without alignment. Subsequently, for... 

This study describes spectrometric monitored kinetic processes to determine the size of citrate-capped Au nanoparticles (Au NPs) based on aggregation induced by L-cysteine (L-Cys) as a molecular linker. The Au NPs association process is thoroughly dependent on pH, concentration and size of nanoparticles. Size dependency of aggregation inspirits to determine the average diameters of Au NPs. For this aim the procedure is achieved in aqueous medium at pH 7 (phosphate buffer), and multivariate data including kinetic spectra of Au NPs are collected during aggregation process. Subsequently partial least squares (PLS) modeling is carried out analyzing the obtained data. The model is built on the... 

A nanowire arrays system consisting of an ordered configuration of Pt, Ni and Co was constructed in single-bath solution through pulse electrodeposition. This structure was evaluated as a potential amperometric non-enzymatic sensor to detect glucose in alkaline solution. We observed a strong and fast amperometric response at low applied potential of 0.4 V vs. SCE over linear ranges of 0-0.2 mM and 0.2-8 mM glucose with sensitivities of 1125 and 333 μA mM-1 cm-2, respectively. We also observed a low detection limit for glucose of 1 μM. Correlation of the electronic and geometric modifications with the electrochemical performance characteristics enhanced catalytic activity of the electrode by... 

A new sensor for simultaneous determination of peroxyacetic acid and hydrogen peroxide using silver nanoparticles (Ag-NPs) as a chromogenic reagent is introduced. The silver nanoparticles have the catalytic ability for the decomposition of peroxyacetic acid and hydrogen peroxide; then the decomposition of them induces the degradation of silver nanoparticles. Hence, a remarkable change in the localized surface plasmon resonance absorbance strength could be observed. Spectra-kinetic approach and artificial neural network was applied for the simultaneous determination of peroxyacetic acid and hydrogen peroxide. Linear calibration graphs were obtained in the concentration range of (8.20×10-5 to... 

This paper proposes and evaluates Low-overhead, Reliable Switch (LRS) architecture to enhance the reliability of Network-on-Chips (NoCs). The proposed switch architecture exploits information and hardware redundancies to eliminate retransmission of faulty flits. The LRS architecture creates a redundant copy of each newly received flit and stores the redundant flit in a duplicated flit buffer that is associated with the incoming channel of the flit. Flit buffers in the LRS are equipped with information redundancy to detect probable bit flip errors. When an error is detected in a flit buffer, its duplicated buffer is used to recover the correct value of the flit. In this way, the propagation... 

Although several analytical models have been proposed for wireless sensor networks (WSNs) with different capabilities, very few of them consider the effect of general service distribution as well as design constraints on network performance. This paper presents a new analytical model to compute message latency in a WSN with loop-free Bellman Ford routing strategy. The model considers limited buffer size for each node using M/G/1/k queuing system. Also, contention probability and resource utilization are suitably modeled. The results obtained from simulation experiments confirm that the model exhibits a high degree of accuracy for various network configurations. © 2009 IEEE