Skip to content
Free planNo credit card requiredStart writing
AI for Research home
AI for Research home
Dashboard
Sign inGet started

Search 250 million research papers

Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.

About 7,601 papers for “parallel performance comparison”

Comparison of Performance of Parallel Computation of CPU Cores on CNN model

Debajit Datta, Dhruv Mittal, Nirmal Philipose Mathew et al. · 2020 International Conference on Emerging Trends in Information Technology and Engineering (ic-ETITE) · 2020

Algorithms such as Convolutional Neural Network used for image classification. When huge convolutional neural networks are used to classify image datasets it takes a lot of time to perform convolution operations, which results in increasing the computational demand of image processing. The better way to accelerate image processing is to parallelizing multiple CPU cores. CPU consists of a few numbers of cores that are optimized for sequential processing. 1GHz. This project aims to compare the performance of parallelized CPUs. To parallelize multicore CPUs Python’s Ray library is being used. Convolutional Neural Network is used as the benchmark image classification algorithm in this project. We have tried to show the speedup in training a model with parallelization of the multicores of a CPU. The comparative study in our work has been carried out amongst three different Convolutional Neural Network models namely, the AlexNet model, the VGG16 model and the model proposed by ‘emmarex’ which we found on the online deep-learning platform, called Kaggle. The dataset used in this project is Plant Disease Image Dataset.

Cited by 11

Two parallel plane jets: Comparison of the performance of three turbulence models

Joseph C. S. Lai, A. Nasr · Proceedings of the Institution of Mechanical Engineers Part G Journal of Aerospace Engineering · 1998

There have been many investigations in the literature to examine the performance of different turbulence models in predicting flow over backward-facing steps where the flow is bounded by solid boundaries. However, the evaluation of different turbulence models in predicting free shear layers with no solid boundaries, such as two parallel plane jets, is limited. In this paper, the velocity field of two parallel plane jets with a small nozzle separation ratio of s/ w = 4:25 determined by laser Doppler anemometry (LDA) is first presented. These experimental results are used to examine the performance of three turbulence models (i.e. k - ɛ, RNG k - ɛ and Reynolds stress) in predicting this flow field. The effects of computational domain size, grid resolution and different discretization schemes on the predictions are discussed. The existence of a recirculation flow region, a merging region and a combined region in the two parallel plane jet configuration has been predicted qualitatively by all three turbulence models. On the other hand, quantitative agreement between predictions and measurements varied by as much as 18 per cent for the merging length while the jet spread in the outer shear layer has been substantially under- predicted by all three models.

Cited by 46

Performance comparison of thrashing control policies for concurrent Mergesorts with parallel prefetching

Kun‐Lung Wu, Philip S. Yu, James Z. Teng · ACM SIGMETRICS Performance Evaluation Review · 1993

We study the performance of various run-time thrashing control policies for the merge phase of concurrent mergesorts using parallel prefetching, where initial sorted runs are stored on multiple disks and the final sorted run is written back to another dedicated disk. Parallel prefetching via multiple disks can be attractive in reducing the response times for concurrent mergesorts. However, severe thrashing may develop due to imbalances between input and output rates, thus a large number of prefetched pages in the buffer can be replaced before referenced. We evaluate through detailed simulations three run-time thrashing control policies: (a) disabling prefetching, (b) forcing synchronous writes and (c) lowering the prefetch quantity in addition to forcing synchronous writes. The results show that (1) thrashing resulted from parallel prefetching can severely degrade the system response time; (2) though effective in reducing the degree of thrashing, disabling prefetching may worsen the response time since more synchronous reads are needed; (3) forcing synchronous writes can both reduce thrashing and improve the response time; (4) lowering the prefetch quantity in addition to forcing synchronous writes is most effective in reducing thrashing and improving the response time.

Cited by 0

Enhancing Application Performance Using Mini-Apps: Comparison of Hybrid Parallel Programming Paradigms

Baurle, Robert, Poteat, Michael, Lawson, Gary et al. · NASA STI Repository (National Aeronautics and Space Administration)

In this work, several mini-apps have been created to enhance a real-world application performance, namely the VULCAN code for complex flow analysis developed at the NASA Langley Research Center. These mini-apps explore hybrid parallel programming paradigms with Message Passing Interface (MPI) for distributed memory access and either Shared MPI (SMPI) or OpenMP for shared memory accesses. Performance testing shows that MPI+SMPI yields the best execution performance, while requiring the largest number of code changes. A maximum speedup of 23X was measured for MPI+SMPI, but only 10X was measured for MPI+OpenMP.

Cited by 0Open access

Performance Comparison for Series and Parallel Modes of a Hybrid Electric Vehicle

Halit Yaşar, Gökhan Canbolat · Sakarya University Journal of Science · 2018

Recently, the studies in the automotive sector has concentrated on hybrid electric vehicle. The better results in fuel economy and vehicle performance values have been achieved with the hybrid electric vehicle technology. In this study, the fuel consumption and vehicle performance values are compared to the series and parallel hybrid electric modes of a conventional street sweeper by using AVL Cruise. Nowadays, there are many driving cycles used in the analysis of fuel consumption and vehicle performans. However the current driving cycles are not appropriate for the vehicles designed for a particular function (carrying, lifting, sweeping etc.) such as street sweepers in terms of velocity and functionality. For this reason, a driving cycle was constructed by taking into consideration the daily tasks of the street sweepers with the help of AVL Cruise. The series and parallel hybrid electric configurations of the street sweeper were analyzed separately. Consequently, the demanded power (kW) from the internal combustion engine and its fuel consumption (L/h) were investiageted under the specific operating conditions in the driving cycle. The simulations show that the series hybrid mode has better fuel economy and vehicle performance.

Cited by 11Open access

PERFORMANCE COMPARISON OF PROPOSED INTERLEAVER WITH DIFFERENT TYPES FOR PARALLEL TURBO CODE

Amjad Ali Jassim, Wael A. H. Hadi · Journal of Engineering and Sustainable Development · 2018

Interleaver type is important in parallel turbo code system. It’s responsible of improve code performance against additive white Gaussian noise AWGN. In this paper, introduce a new interleaver with its mathematical model. Study its performance as comparison with different interleavers, including random interleaver, QPP quadratic primitive polynomial interleaver as standard of LTE, chaotic interleaver and finally 2-D interleaver for burst error environment, Study the performance as simulation. The system takes in consideration also different modulation schemes including QPSK, 16-QAM and 64-QAM. Result comparison between different types summarized in performance BER curves and tables.

Cited by 2Open access

Performance comparison for close parallel lines and lines crossing at small angles.

Benjamin Eltzner (742711), Carina Wollnik (742712), Carsten Gottschlich (643445) et al. · Figshare · 2015

<p>Showing details from the simulated test image displayed in <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0126346#pone.0126346.g011" target="_blank">Fig 11</a>. The columns show, from left to right, the original detail, the results of the FS, the results of the eLoG method, and the results of CID. Green pixels are false positives detected by the method, yellow are correctly identified pixels and red are missed pixels as in <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0126346#pone.0126346.g009" target="_blank">Fig 9</a>. In the details shown the FS and eLoG method clearly outperform CID, which does poorly at detecting close parallel lines, as shown in the first three rows of the figure. Also for crossing lines as displayed in the fourth and fifth row, CID performs poorly. The eLoG method shows some weaknesses for close parallel lines and overshoots at the endings of lines. Overall, its line pixel detection capabilities in these synthetic test cases are comparable to those of the FS.</p>

Cited by 0Open access

Comparison of Serial and Parallel Programming Performance in Outlier Detection with DBSCAN Algorithm

Hüseyin YAŞAR, Mehmet ALBAYRAK · Bilecik Şeyh Edebali Üniversitesi Fen Bilimleri Dergisi · 2020

With the introduction of computers into our lives, digital data sizes are increasing gradually. Non-standard values (outliers) which behave differently from the others can be found in these data produced in the digital world. Detection of these values, especially in big data sets; has great importance in fields such as security, insurance, finance, medicine and genetics. Clustering methods of data mining techniques are frequently used in outlier detection in big data sets. Density based DBSCAN (Density-based spatial clustering of applications with noise) algorithm from clustering algorithms which are sensitive to noisy and outlier values is one of the most important methods in outlier detection. In this study, an application was developed using DBSCAN algorithm in C# programming language for the detection of outliers. In the developed application, 2 data sets with different data numbers were examined and analyzed. For the shortest possible data analysis time, serial and parallel programming techniques were used separately. In order to shorten the analysis time of big data sets, parallel class members in TPL (Task Parallel Library) provided with .Net 4.0 were used. In series of analysis of data sets, it was seen that DBSCAN algorithm produces more accurate results and is more practicable than other selected algorithms in terms of outlier detection. When considered in terms of computing performance, parallel programming has become more efficient as the number of data increases.

Cited by 1Open access

The Performance Comparison Analysis on Hydraulic Muffler of Quality Room with Parallel Line

Chang Ji Shan, Yi Bian, Nan Xu · Applied Mechanics and Materials · 2014

Though establishing physical modeling of porous hydraulic mufflers, it could derive the transfer matrix. Porous hydraulic muffler attenuation characteristic curve is obtained by MATLAB programming. Through the analysis of characteristic curve, the result showed that the porous hydraulic muffler can effectively absorb fluid pulsation and it is concluded that the change of the structure parameters affect on the attenuation characteristics. It can provide a reference for related research platform.

Cited by 0

Comparison of PEMFC performance with parallel-serpentine flow field under various geometrical conditions

Mengyuan Lei, Kai Wang, Jifei Liu et al. · 2023

To study the effect of the number of parallel serpentine flow channels on the performance of proton exchange membrane fuel cell (PEMFC), a multi physical field simulation software COMSOL was used to simulate the flow channel structures with different number of channels. The effects of the number of channels on the pressure drop and fluid mass flow rate in the channels were analyzed. The results show that with the number of flow channels increases, the drainage capacity and the power density also increases, but at the same time, it also increases the pressure drop inside the flow channel.

Cited by 0

Performance comparison between channel estimation algorithms for TD-SCDMA parallel-interference-cancellation receiver

Gang Wu, Wenhua Dai, Youxi Tang et al. · 2003

The channel estimation problem for TD-SCDMA parallel interference cancellation (PIC) receiver is considered. Accurate information of channel impulsive response (CIR) is the key to the performance of the receivers. A successive interference-cancellation based estimation method is proposed, and it is compared with the conventional correlation method and the joint channel estimation method by link level simulation with the COSSAP platform. The simulation results show that the suggested channel estimator can provide more accurate estimates than the conventional method under a high-speed vehicular channel.

Cited by 0

A comparison of the performance of three parallel architectures for matrix-vector multiplication

Codenotti B, Puglisi C. · ISTI Open Portal · 1988

We present the paralle1 implementation of matrix-vector multiplication on a binary tree of a small number of processor, whose leaves are connected to loca1 memories, each containing one column of the matrix. The performance attained can be favourably compared with the one of the mesh of trees and the linear array, each formed by the same number of processors.

Cited by 0Open access