MATLAB Distributed Computing Server Research Articles

Big data analytics in Industry 4.0 ecosystems

Big data analytics in Industry 4.0 ecosystems

MCBTE: A variance-reduced Monte Carlo solution of the linearized Boltzmann transport equation for phonons

MCBTE solves the linearized Boltzmann transport equation for phonons in three-dimensions using a variance-reduced Monte Carlo solution approach. The algorithm is suited for both transient and steady-state analysis of thermal transport in structured materials with size features in the nanometer to hundreds of microns range. The code is portable and integrated with both first-principles density functional theory calculations and empirical relations for the input of phonon frequency, group velocity, and mean free path required for calculating the thermal properties. The program outputs space- and time-resolved temperature and heat flux for the transient study. For the steady-state simulations, the frequency-resolved contribution of phonons to temperature and heat flux is written to the output files, thus allowing the study of cumulative thermal conductivity as a function of phonon frequency or mean free path. We provide several illustrative examples, including ballistic and quasi-ballistic thermal transport, the thermal conductivity of thin films and periodic nanostructures, to demonstrate the functionality and to benchmark our code against available theoretical/analytical/computational results from the literature. Moreover, we parallelize the code using the Matlab Distributed Computing Server, providing near-linear scaling with the number of processors. Program summaryProgram Title:MCBTECPC Library link to program files:https://doi.org/10.17632/rzn86w7t3p.1Developer's repository link:https://github.com/abhipath90/MCBTECode Ocean capsule:https://codeocean.com/capsule/7196257Licensing provisions: GPLv3Programming language: MATLABNature of problem: Calculation of time- and space-dependent temperature and heat flux profiles, and frequency-resolved effective thermal conductivity in structured systems where heat is carried by phonons.Solution method: Solution of linearized Boltzmann transport equation for phonons, variance-reduced Monte Carlo approach.Additional comments including restrictions and unusual features: Runtime about 1 to 10 hours on a personal computer.

Numerical simulation of DIC drying process on matlab distributed computing server

<span>Instant controlled pressure drop, also known as DIC, is one of the drying techniques that has been used for texturing, extracting and drying various food products. Mathematical model has been used to explain the drying process, although most of the studies focused on the statistical regression model approach. Due to the limitations of regression model, which neglects the fundamental of dehydration process, t</span><span>his paper presents the development of mathematical models to detect, solve and visualize the three-dimensional (3D) heat and mass transfer in DIC drying process. </span><span>Finite Difference Method (FDM) with central difference formula is used to discretize the mathematical models. A large sparse of system of linear equation (SLE), which represents the actual drying process, is solved by using some numerical methods, such as Jacobi (JB), Red Black Gauss Seidel (RBGS), Alternating Group Explicit with Douglas (AGED) and Brian (AGEB) methods. Based on the numerical results, high execution time and high computational complexity have been shown. In order to reduce the execution time and computational complexity, the parallel algorithm based on domain decomposition technique has been implemented on the MATLAB Distributed Computing Server (MDCS). The parallel algorithm of the numerical methods was evaluated and compared based on the parallel performance metrics: execution time, speed up, efficiency, effectiveness, temporal performance and granularity. From the parallel performance metrics, it was found that the PAGEB approach had better performance, followed by PAGED, PRBGS and PJB methods.</span>

An efficient Hadoop‐based brain tumor detection framework using big data analytic

AbstractThe exponential increase of brain MR image data in the medical imaging field requires faster and accurate segmentation of tumor. The computer aided detection systems acting as a second option to experts, radiologists, and surgeons needs to be swift enough to handle parallelism. However, handling of massive MR data for segmentation with high accuracy and low processing time is significant concern of any framework. In this article, distributed platforms for brain tumor segmentation using hybrid weighted fuzzy approach integrated with Matlab Distributed Computing Server and Hadoop has been proposed. The approach is based on the fuzzification of the pixel values to achieve more meaningful clusters by grouping of large data into similar clusters. The article focuses on analyzing the performance of varying sized data sets using hybrid fuzzy clustering in MapReduce on Hadoop to deal with huge MR brain data cross clusters of commodity computers. For experimentation varying size of DICOM data set is processed through different number of clusters to compare the read, write, and processing time on each node. The read and write operation time elevates as the data size increasing is floated to multinode. However, the processing time of the proposed approach turns to be 35 min on single, whereas 3‐node clusters process the same data set (215 MB) in 3.4 min. Furthermore, increasing the data set to 7.3 GB the 3‐node cluster performs in 235.4 min which is greatly reduced from single node processing time of 2085.2 min.

Parallel Mutant Execution Techniques in Mutation Testing Process for Simulink Models

Mutation testing – a fault-based technique for software testing – is a computationally expensive approach. One of the powerful methods to improve the performance of mutation without reducing effectiveness is to employ parallel processing, where mutants and tests are executed in parallel. This approach reduces the total time needed to accomplish the mutation analysis. This paper proposes three strategies for parallel execution of mutants on multicore machines using the Parallel Computing Toolbox (PCT) with the Matlab Distributed Computing Server. It aims to demonstrate that the computationally intensive software testing schemes, such as mutation, can be facilitated by using parallel processing. The experiments were carried out on eight different Simulink models. The results represented the efficiency of the proposed approaches in terms of execution time during the testing process.

Simulating Electrohydrodynamic Ion-Drag Pumping on Distributed Parallel Computing Systems

Objectives: This paper aims to simulate EHD ion-drag pumping model using Finite Difference Method (FDM) and to apply the idea of parallelism to reduce the computational time. Methods: The numerical simulation of EHD ion-drag pumping plays an important part not only to understand the different working principles but also enables to model the designs with better performance. Since the performance of EHD pumps depends on the shapes and geometries of the actuator electrodes, therefore the variation in the geometric dimensions of the electrodes require dense and fine meshes for numerical solution. Consequently, the numerical simulations take unacceptably more execution time on sequential computers. For that reason, a Data Parallel Algorithm for EHD model (DPA-EHD) is designed. To implement the parallel algorithm a distributed parallel computing system using MATLAB Distributed Computing Server (MDCS) is configured. The computational time and speedup with respect to the different number of processors is evaluated. Findings: This results show that the parallel algorithm for EHD simulations may provide 4.14 times more speedup over sequential algorithm for large grid sizes. Improvements: This study shows the feasibility of using the parallelism to reduce the computational time in the EHD model enabling to simulate the micropumps with very small dimensions of electrodes.

PARALLEL COMPUTING OF NUMERICAL SCHEMES AND BIG DATA ANALYTIC FOR SOLVING REAL LIFE APPLICATIONS

This paper proposed the several real life applications for big data analytic using parallel computing software. Some parallel computing software under consideration are Parallel Virtual Machine, MATLAB Distributed Computing Server and Compute Unified Device Architecture to simulate the big data problems. The parallel computing is able to overcome the poor performance at the runtime, speedup and efficiency of programming in sequential computing. The mathematical models for the big data analytic are based on partial differential equations and obtained the large sparse matrices from discretization and development of the linear equation system. Iterative numerical schemes are used to solve the problems. Thus, the process of computational problems are summarized in parallel algorithm. Therefore, the parallel algorithm development is based on domain decomposition of problems and the architecture of difference parallel computing software. The parallel performance evaluations for distributed and shared memory architecture are investigated in terms of speedup, efficiency, effectiveness and temporal performance.

Spectral Clustering Algorithm: MATLAB PCT-Based Parallel Design and Implementation

Spectral clustering algorithms inevitable exist computational time and memory use problems for large-scale spectral clustering, owing to compute-intensive and data-intensive. We analyse the time complexity of constructing similarity matrix, doing eigendecomposition and performing k-means and exploiting SPMD parallel structure supported by MATLAB Parallel Computing Toolbox (PCT) to decrease eigendecomposition computational time. We propose using MATLAB Distributed Computing Server to parallel construct similarity matrix, whilst using t-nearest neighbors approach to reduce memory use. Ultimately, we present clustering time, clustering quality and clustering accuracy in the experiments.

Development of Priority Oriented Scheduling Method to Increase the Reliability of Manufacturing Systems

Scheduling occurs in every organization without considering the nature of its activities. In this regard, numerous scholars have attempted to schedule via divergent methods including classical scheduling, genetic algorithm, neural network and fuzzy logic and so on. Studies in manufacturing scheduling mostly deal with priority rules without any consideration of the system states. An appropriate scheduling leads to significant enhancement of fairness in job scheduling. The term fairness can be transformed into a specific selection of job weights. There is no method of scheduling in which priority, time action (duration) and time stamp of jobs have simultaneously been considered. But the proposed method of scheduling can enhance the efficiency and reliability of manufacturing systems. To fulfill this target, first and foremost, the normalize method should be performed. This method allows data (time stamp, time action, priority) of jobs on different scales to be compared by bringing them to a common scale. Secondly, the jobs should be arranged based on three criteria which are priority, time action and time stamp. This sorting algorithm is programmed via MATLAB Distributed Computing Server (DCS) software. Eventually, to evaluate the proposed method of scheduling, simulation is operated. The simulated algorithm shows that applying the proposed method of scheduling increases the efficiency of simulated scheduler in comparison with applying the common method of scheduling. Besides the mentioned simulated algorithm, there is a mathematical proof to show the enhancement of reliability.