Optimizing Memory Access Efficiency in CUDA Kernel via Data Layout Technique

Over the past decade, Graphics Processing Units (GPUs) have revolutionized high-performance computing, playing pivotal roles in advancing fields like IoT, autonomous vehicles, and exascale computing. Despite these advancements, efficiently programming GPUs remains a daunting challenge, often relying on trial-and-error optimization methods. This paper introduces an optimization technique for CUDA programs through a novel Data Layout strategy, aimed at restructuring memory data arrangement to significantly enhance data access locality. Focusing on the dynamic programming algorithm for chained matrix multiplication—a critical operation across various domains including artificial intelligence (AI), high-performance computing (HPC), and the Internet of Things (IoT)—this technique facilitates more localized access. We specifically illustrate the importance of efficient matrix multiplication in these areas, underscoring the technique’s broader applicability and its potential to address some of the most pressing computational challenges in GPU-accelerated applications. Our findings reveal a remarkable reduction in memory consumption and a substantial 50% decrease in execution time for CUDA programs utilizing this technique, thereby setting a new benchmark for optimization in GPU computing.

Pattern recognition and data mining software based on artificial neural networks applied to proton transfer in aqueous environments

In computational physics proton transfer phenomena could be viewed as pattern classification problems based on a set of input features allowing classification of the proton motion into two categories： transfer ＇occurred＇ and transfer ＇not occurred＇. The goal of this paper is to evaluate the use of artificial neural networks in the classification of proton transfer events, based on the feed-forward back propagation neural network, used as a classifier to distinguish between the two transfer cases. In this paper, we use a new developed data mining and pattern recognition tool for automating, controlling, and drawing charts of the output data of an Empirical Valence Bond existing code. The study analyzes the need for pattern recognition in aqueous proton transfer processes and how the learning approach in error back propagation （multilayer perceptron algorithms） could be satisfactorily employed in the present case. We present a tool for pattern recognition and validate the code including a real physical case study. The results of applying the artificial neural networks methodology to crowd patterns based upon selected physical properties （e.g., temperature, density） show the abilities of the network to learn proton transfer patterns corresponding to properties of the aqueous environments, which is in turn proved to be fully compatible with previous proton transfer studies.

Detecting network intrusions by data mining and variable-length sequence pattern matching

Anomaly detection has been an active research topic in the field of network intrusion detection for many years. A novel method is presented for anomaly detection based on system calls into the kernels of Unix or Linux systems. The method uses the data mining technique to model the normal behavior of a privileged program and uses a variable-length pattern matching algorithm to perform the comparison of the current behavior and historic normal behavior, which is more suitable for this problem than the fixed-length pattern matching algorithm proposed by Forrest et al. At the detection stage, the particularity of the audit data is taken into account, and two alternative schemes could be used to distinguish between normalities and intrusions. The method gives attention to both computational efficiency and detection accuracy and is especially applicable for on-line detection. The performance of the method is evaluated using the typical testing data set, and the results show that it is significantly better than the anomaly detection method based on hidden Markov models proposed by Yan et al. and the method based on fixed-length patterns proposed by Forrest and Hofmeyr. The novel method has been applied to practical hosted-based intrusion detection systems and achieved high detection performance.

Exploring Temporal Activity Patterns of Urban Areas Using Aggregated Network-driven Mobile Phone Data:A Case Study of Wuhu,China

The increasing availability of data in the urban context(e.g.,mobile phone,smart card and social media data)allows us to study urban dynamics at much finer temporal resolutions(e.g.,diurnal urban dynamics).Mobile phone data,for instance,are found to be a useful data source for extracting diurnal human mobility patterns and for understanding urban dynamics.While previous studies often use call detail record(CDR)data,this study deploys aggregated network-driven mobile phone data that may reveal human mobility patterns more comprehensively and can mitigate some of the privacy concerns raised by mobile phone data usage.We first propose an analytical framework for characterizing and classifying urban areas based on their temporal activity patterns extracted from mobile phone data.Specifically,urban areas’diurnal spatiotemporal signatures of human mobility patterns are obtained through longitudinal mobile phone data.Urban areas are then classified based on the obtained signatures.The classification provides insights into city planning and development.Using the proposed framework,a case study was implemented in the city of Wuhu,China to understand its urban dynamics.The empirical study suggests that human activities in the city of Wuhu are highly concentrated at the Traffic Analysis Zone(TAZ)level.This large portion of local activities suggests that development and planning strategies that are different from those used by metropolitan Chinese cities should be applied in the city of Wuhu.This article concludes with discussions on several common challenges associated with using network-driven mobile phone data,which should be addressed in future studies.

An Efficient Outlier Detection Approach on Weighted Data Stream Based on Minimal Rare Pattern Mining

The distance-based outlier detection method detects the implied outliers by calculating the distance of the points in the dataset, but the computational complexity is particularly high when processing multidimensional datasets. In addition, the traditional outlier detection method does not consider the frequency of subsets occurrence, thus, the detected outliers do not fit the definition of outliers (i.e., rarely appearing). The pattern mining-based outlier detection approaches have solved this problem, but the importance of each pattern is not taken into account in outlier detection process, so the detected outliers cannot truly reflect some actual situation. Aimed at these problems, a two-phase minimal weighted rare pattern mining-based outlier detection approach, called MWRPM-Outlier, is proposed to effectively detect outliers on the weight data stream. In particular, a method called MWRPM is proposed in the pattern mining phase to fast mine the minimal weighted rare patterns, and then two deviation factors are defined in outlier detection phase to measure the abnormal degree of each transaction on the weight data stream. Experimental results show that the proposed MWRPM-Outlier approach has excellent performance in outlier detection and MWRPM approach outperforms in weighted rare pattern mining.

Local binary pattern-based reversible data hiding

A novel local binary pattern-based reversible data hiding(LBP-RDH)technique has been suggested to maintain a fair symmetry between the perceptual transparency and hiding capacity.During embedding,the image is divided into various 3×3 blocks.Then,using the LBP-based image descriptor,the LBP codes for each block are computed.Next,the obtained LBP codes are XORed with the embedding bits and are concealed in the respective blocks using the proposed pixel readjustment process.Further,each cover image(CI)pixel produces two different stego-image pixels.Likewise,during extraction,the CI pixels are restored without the loss of a single bit of information.The outcome of the proposed technique with respect to perceptual transparency measures,such as peak signal-to-noise ratio and structural similarity index,is found to be superior to that of some of the recent and state-of-the-art techniques.In addition,the proposed technique has shown excellent resilience to various stego-attacks,such as pixel difference histogram as well as regular and singular analysis.Besides,the out-off boundary pixel problem,which endures in most of the contemporary data hiding techniques,has been successfully addressed.

Wind Speed and Altitude Dependent AMDAR Observational Error and Its Impacts on Data Assimilation and Forecasting

Aircraft Meteorological Data Relay(AMDAR)observations have been widely used in numerical weather prediction(NWP)because of its high spatiotemporal resolution.The observational error of AMDAR is influenced by aircraft flight altitude and atmospheric condition.In this study,the wind speed and altitude dependent observational error of AMDAR is estimated.The statistical results show that the temperature and the observational error in wind speeds slightly decrease as altitude increases,and the observational error in wind speed increases as wind speed increases.Pseudo single AMDAR observation assimilation tests demonstrate that the wind speed and altitude dependent observational error can provide more reasonable analysis increment.Furthermore,to assess the performance of wind speed and altitude dependent observational error on data assimilation and forecasting,two-month 3-hourly cycling data assimilation and forecast experiments based on the Weather Research and Forecasting Model(WRF)and its Data Assimilation system(WRFDA)are performed for the period during 1 September-31 October,2017.The results of the two-month 3-hourly cycling experiments indicate that new observational error improves analysis and forecast of wind field and geo-potential height,and has slight improvements on temperature.The Fractions Skill Score(FSS)of the 6-h accumulated precipitation shows that new wind speed and altitude dependent observational error leads to better precipitation forecast skill than the default observational error in the WRFDA does.

Multidimensional Visualization of Bikeshare Travel Patterns Using a Visual Data Mining Technique: Data Cubes

In order to explore the travel characteristics and space-time distribution of different groups of bikeshare users,an online analytical processing(OLAP)tool called data cube was used for treating and displaying multi-dimensional data.We extended and modified the traditionally threedimensional data cube into four dimensions,which are space,date,time,and user,each with a user-specified hierarchy,and took transaction numbers and travel time as two quantitative measures.The results suggest that there are two obvious transaction peaks during the morning and afternoon rush hours on weekdays,while the volume at weekends has an approximate even distribution.Bad weather condition significantly restricts the bikeshare usage.Besides,seamless smartcard users generally take a longer trip than exclusive smartcard users;and non-native users ride faster than native users.These findings not only support the applicability and efficiency of data cube in the field of visualizing massive smartcard data,but also raise equity concerns among bikeshare users with different demographic backgrounds.

A Test Pattern Identification Algorithm and Its Application to CINRAD/SA(B) Data

A variety of faulty radar echoes may cause serious problems with radar data applications,especially radar data assimilation and quantitative precipitation estimates.In this study,＂test pattern＂ caused by test signal or radar hardware failures in CINRAD （China New Generation Weather Radar） SA and SB radar operational observations are investigated.In order to distinguish the test pattern from other types of radar echoes,such as precipitation,clear air and other non-meteorological echoes,five feature parameters including the effective reflectivity data percentage （Rz）,velocity RF （range folding） data percentage （RRF）,missing velocity data percentage （RM）,averaged along-azimuth reflectivity fluctuation （RNr,z） and averaged along-beam reflectivity fluctuation （RNa,z） are proposed.Based on the fuzzy logic method,a test pattern identification algorithm is developed,and the statistical results from all the different kinds of radar echoes indicate the performance of the algorithm.Analysis of two typical cases with heavy precipitation echoes located inside the test pattern are performed.The statistical results show that the test pattern identification algorithm performs well,since the test pattern is recognized in most cases.Besides,the algorithm can effectively remove the test pattern signal and retain strong precipitation echoes in heavy rainfall events.

Detection of Alzheimer’s disease onset using MRI and PET neuroimaging:longitudinal data analysis and machine learning

The scientists are dedicated to studying the detection of Alzheimer’s disease onset to find a cure, or at the very least, medication that can slow the progression of the disease. This article explores the effectiveness of longitudinal data analysis, artificial intelligence, and machine learning approaches based on magnetic resonance imaging and positron emission tomography neuroimaging modalities for progression estimation and the detection of Alzheimer’s disease onset. The significance of feature extraction in highly complex neuroimaging data, identification of vulnerable brain regions, and the determination of the threshold values for plaques, tangles, and neurodegeneration of these regions will extensively be evaluated. Developing automated methods to improve the aforementioned research areas would enable specialists to determine the progression of the disease and find the link between the biomarkers and more accurate detection of Alzheimer’s disease onset.

Spatial-Temporal Features of Wuhan Urban Agglomeration Regional Development Pattern—Based on DMSP/OLS Night Light Data

Based on the night light data, urban area data, and economic data of Wuhan Urban Agglomeration from 2009 to 2015, we use spatial correlation dimension, spatial self-correlation analysis and weighted standard deviation ellipse to identify the general characteristics and dynamic evolution characteristics of urban spatial pattern and economic disparity pattern. The research results prove that: between 2009 and 2013, Wuhan Urban Agglomeration expanded gradually from northwest to southeast and presented the dynamic evolution features of “along the river and the road”. The spatial structure is obvious, forming the pattern of “core-periphery”. The development of Wuhan Urban Agglomeration has obvious imbalance in economic geography space, presenting the development tendency of “One prominent, stronger in the west and weaker in the east”. The contract within Wuhan Urban Agglomeration is gradually decreased. Wuhan city and its surrounding areas have stronger economic growth strength as well as the cities along The Yangtze River. However, the relative development rate of Wuhan city area is still far higher than other cities and counties.

Distribution Pattern of Urban Housing Price Based on Multiple Big Data:A Case Study of Beijing,Shanghai,Guangzhou,and Wuhan

The housing price has been paid close attention by people in all walks of life,and the development of big data provides a new data environment for the study of urban housing price.Housing price data of four national central cities (Beijing,Shanghai,Guangzhou and Wuhan) are taken as research samples.With the help of software GIS,exploratory spatial data analysis method is used to depict the spatial distribution pattern of urban housing price,and commonness and difference of spatial distribution of housing price are explored.The conclusions are as below:①regional imbalance of housing price in national central cities is significant.②Spatial distribution of urban housing price in Beijing,Shanghai,Guangzhou and Wuhan presents a polycentric pattern,and there is obvious spatial agglomeration.③The internal change of housing price in different cities has significant spatial difference.Beijing,Shanghai,Guangzhou and Wuhan are taken as typical city samples for research,with reference and practical significance,which could help to effectively predict spatial development trend of housing prices in other first and second tier cities.The research aims to provide certain reference for the government implementing real estate control policies according to local conditions,project location and reasonable pricing of real estate developers.

Research on the Modern Precision E-commerce Marketing Model under the Big Data and Pattern Recognition Background

In this paper, we conduct research on the modern precision e-commerce marketing model under the big data and pattern recognition background. Large amount of consumption data provides the electricity enterprises grasp the user consumption pattern and the basis of the electric business enterprise through the use of big data can be personalized, accurate and intelligent advertising push service, service mode for the creation of more interesting and effective. Under this basis, electricity companies can also pass the assurance of pair of big data, looking for better increase user stickiness, development of new products and services, the ways and methods to reduce operational costs and accordingly, we propose the novel perspectives on the corresponding issues for the systematic level enhancement that provides the novel methodology of precision e-commerce marketing.

5-33 Design Pattern in Data Acquisition Monitor System for HIMM Power Supply

In Heavy-Ion Medical Machine (HIMM), it is very necessary to monitor the power supply's current value bythe data acquisition The hardware of the data acquisition monitor system is NI PXIe-6366 DAQ card, which has16-bits resolution, 8 differential channels, 2MS/s per channel's sample rate, and also has a good synchronizationperformance. The hardware architecture is shown in Fig. 1.

Relational Logging Design Pattern

Observability and traceability of developed software are crucial to its success in software engineering.Observability is the ability to comprehend a system’s internal state from the outside.Monitoring is used to determine what causes system problems and why.Logs are among the most critical technology to guarantee observability and traceability.Logs are frequently used to investigate software events.In current log technologies,software events are processed independently of each other.Consequently,current logging technologies do not reveal relationships.However,system events do not occur independently of one another.With this perspective,our research has produced a new log design pattern that displays the relationships between events.In the design we have developed,the hash mechanism of blockchain technology enables the display of the logs’relationships.The created design pattern was compared to blockchain technology,demonstrating its performance through scenarios.It has been determined that the recommended log design pattern outperforms blockchain technology in terms of time and space for software engineering observability and traceability.In this context,it is anticipated that the log design pattern we provide will strengthen the methods used to monitor software projects and ensure the traceability of relationships.

Numerical Comparison of Shapeless Radial Basis Function Networks in Pattern Recognition

This work focuses on radial basis functions containing no parameters with themain objective being to comparatively explore more of their effectiveness.For this,a total of sixteen forms of shapeless radial basis functions are gathered and investigated under the context of the pattern recognition problem through the structure of radial basis function neural networks,with the use of the Representational Capability(RC)algorithm.Different sizes of datasets are disturbed with noise before being imported into the algorithm as‘training/testing’datasets.Each shapeless radial basis function is monitored carefully with effectiveness criteria including accuracy,condition number(of the interpolation matrix),CPU time,CPU-storage requirement,underfitting and overfitting aspects,and the number of centres being generated.For the sake of comparison,the well-known Multiquadric-radial basis function is included as a representative of shape-contained radial basis functions.The numerical results have revealed that some forms of shapeless radial basis functions show good potential and are even better than Multiquadric itself indicating strongly that the future use of radial basis function may no longer face the pain of choosing a proper shape when shapeless forms may be equally(or even better)effective.