Stroke commonly leads to long-term gait impairments, underscoring the need for objective and continuous functional assessment during rehabilitation. This study employs machine learning methods to assess a smart insole-based system in stroke gait recognition. Data were collected from stroke survivors and healthy control participants during Walk and Timed-Up-and-Go tasks. After preprocessing, group differences were quantified using Hedges' g, and multiple machine learning models were applied to classify two participant groups. Support Vector Machine and KNN achieved the best performance, with accuracies of 0.88. The results demonstrate that sensor-based gait features can be used to distinguish stroke gait patterns from control gait patterns, highlighting the potential of this approach for future homebased monitoring and personalised rehabilitation.

Real-time monitoring is essential in industrial applications such as tool wear detection and temperature compensation, where maintaining process stability and ensuring product quality are critical. To address the challenges caused by heterogeneous data sources and device types, this paper proposes a real-time edge computing system based on the Intel Edge Insights for Industrial (EII) platform. The system utilizes multiple AI Neural Compute Sticks (NCS) to enable accelerated and parallel inference for multiple tasks. The system integrates two AI models in separate containers. One model performs tool wear detection using a U-Net segmentation network, and the other conducts temperature compensation through a one-dimensional convolutional neural network. Each container operates on an independent NCS to support efficient and isolated task execution. Data communication is managed by the EII Message Bus, stored in InfluxDB, and visualized using Grafana, enabling complete data flow from acquisition to real-time presentation. Experimental results demonstrate that the system can concurrently handle multiple monitoring tasks with low latency and reliable performance. The proposed architecture enhances operational efficiency, supports intelligent automation, and provides a scalable solution for smart manufacturing environments.

Detection and recognition of traffic signs are two analytic processes in vehicular systems that contribute to increasing driver safety, warning, and preventing collisions by improving drivers’ focus and awareness. They are also crucial for developing self-driving vehicles that can sense the environment through a camera eye and understand the road restrictions. Convolutional Neural Networks (CNNs) play an essential role in both processes by finding the traffic sign objects on acquired images or video frames and recognizing their meaning. However, CNN architectures running on vehicles need minimization, ensuring satisfactory performance and efficient operation with decreased computational resources. In this paper, we investigate two CNN architectures for traffic sign detection and two architectures for traffic sign recognition. Our experiments confirm that light models for both processes can successfully perform the achieved tasks, reaching effectiveness close to the complex models reported in the scientific literature.

Fuzzy data processing enables data enrichment and increases data interpretation in industrial environments. In the cloud-based IoT data ingestion pipelines, fuzzy data processing can be implemented in several locations, closer to the IoT events gateways, stream processors, or the persistence layer before the data is visualized. Since Automated Guided Vehicles (AGV)-enabled manufacturing can produce vast amounts of data, the decision on the placement of the fuzzy data processing can be important for secondary processes performed on the enriched data, like the predictive maintenance inferencing. In this paper, we analyze two locations of fuzzy data processing in the cloud-based environment built for monitoring AGVs in smart factories - by formulating fuzzy queries against data streams on stream processing units and data at rest in a database. The querying scenarios cover fuzzy filtering with simple and complex criteria, fuzzy filtering through assignment to a linguistic variable, and joining data streams by representing joining attributes as fuzzy numbers. The experimental results show that querying the data stream can be more efficient and profitable in the scalable environment of many AGVs. However, the enrichment provided for the data at rest is also beneficial when gathering data for building future predictive maintenance models.

Proteins are complex biological information granules that play a crucial role in various cellular processes within living organisms. Processing 3D protein structures, which are the most informative from the biological point of view, is both intricate and time-consuming. In particular, performing 3D protein structure searches against large protein datasets involves identifying similarities and conducting structural alignments across numerous molecules (granules). This task demands advanced methods for matching identical and similar regions within protein structures and substantial computational resources to handle large collections of macromolecular data efficiently. In this paper, we present our parallel implementation of scalable 3D structural alignment on the Apache Spark big data platform. We describe a customized approach that leverages Spark data transformations within the data processing pipeline for the alignment process. Our experimental results demonstrate that this solution, tightly integrated with the Spark processing model, is both efficient and scalable, even with the increasing volume of protein structure data.

AI-driven tool wear monitoring models playa crucial role in the manufacturing sector by accurately forecasting and identifying tool degradation. This capability enables the reduction of downtime and ensures the maintenance of cutting quality. Nevertheless, these models encounter obstacles in noisy manufacturing settings, where environmental variables may disrupt sensor data, impacting the precision of wear predictions. As a result, this article presents a novel UNet-based noise reduction model designed to eliminate diverse environmental noise from cutting sounds. This model is trained using hybrid signals in both the time-domain and frequency domain as part of the loss function. It enables the model to capture both temporal and spectral characteristics of the data, allowing for a more comprehensive representation of the signal's behavior. Experiments show that the Signal-to-Noise Ratio (SNR) can be effectively increased by over 3dB compared to the baseline across various cutting workpieces. Additionally, the proposed method exhibits superior robustness against various types and levels of noise. The results demonstrate that the quality of the cutting sound can be enhanced by over 7dB and 4dB respectively, following the application of the noise reduction technique.

As the COVID-19 pandemic continues, a growing number of recovered patients report persistent symptoms such as fatigue, muscle weakness, sleep issues, anxiety, and depression, lasting months or even over a year. Severe cases often show significant lung damage and sometimes reduced kidney function. This study examines a dataset from recovered COVID-19 patients, using machine learning to assess the likelihood of developing Post-COVID-19 conditions. We applied several models, including XGBoost, Decision Trees, and Random Forest, to predict outcomes based on data from a specific hospital. Our approach included detailed data preprocessing-filling in missing values, feature engineering, and standardizing data to improve model accuracy and applicability. Results showed the Random Forest model as the most accurate, demonstrating the power of machine learning in making precise predictions from complex health data. Feature importance analysis revealed critical factors predicting Post-COVID-19 conditions, offering vital guidance for healthcare professionals in managing recovered patients.

Automated monitoring of tool wear is crucial for maintaining product quality. Furthermore, implementing AI techniques for real-time tool monitoring involves not only developing models but also managing their versions, avoiding the issue of models becoming less accurate as the properties of the machinery change over time. Consequently, this study develops a tool wear prediction system integrated with an artificial intelligent (AI) model management platform. First, this system uses various machine learning models to extract diverse signal features from sensor fusion, thereby boosting the accuracy of tool wear prediction. Secondly, the AI Models Management Platform comprises the C# programming language, Neural Networks Processing Unit (NPU) board, and Docker on both user and server sides, enhancing industrial processes and enabling real-time analysis of sensor data. According to these results, the Ensemble Learning method within the machine learning model demonstrates superior performance, yielding an average root mean squared error (RMSE) of$\mathbf{0.000185}\ mm^{2}$. Additionally, AI model management platform efficiently handle various model versions and streamline data training processes, empowering users to select suitable models and thereby enhancing system robustness.

In industrial settings, querying data streams from Internet of Things (IoT) devices benefits from utilizing elastic criteria to enhance the interpretability of the current state of the monitored environment. Fuzzy sets provide this elasticity, enabling the aggregation and representation of similar values in a human-comprehensible manner. However, many sensor signals exhibit temporal oscillations, leading to varying interpretations of the signal based on its current trend (rising or falling). This hysteresis in signal (and subsequently of the production device) interpretation inspired us to introduce this phenomenon into data stream processing, resulting in the novel concept of hysteretic fuzzy sets. This paper demonstrates how fuzzy searching and grouping can be applied to IoT sensor signals in flexible Big Data stream processing on Apache Kafka. We illustrate the impact of data stream querying with KSQL queries involving fuzzy sets (encompassing fuzzy filtering of data stream events, fuzzy transformation of data stream attributes, fuzzy grouping, and joining) on the flexibility of executed operations and computational resources utilized by the Kafka processing engine. Finally, our experiments with hysteretic fuzzy sets while analyzing sensor signals in power plants demonstrate that this novel approach effectively reduces the number of alarms while monitoring the state of the production machine.

Macular holes (MHs) can be either idiopathic or secondary to a result of concurrent or previous pathology such as ocular inflammation, trauma or surgery. Idiopathic macular holes may eventually impair the life quality and self-care capability of patients. In clinical, the optical coherence tomography (OCT) images of macular holes can be divided into 4 stages based upon the size of macular hole. The size of macular hole is inversely influencing surgical success rate and visual outcomes. Minimizing human judgment errors in the classification of MHs and enhancing the overall quality of classification are critical objectives. In this paper, we develop a deep learning algorithm to detect MHs on OCT images and also propose an automatic algorithm to measure the size of MH. The MH detection algorithm demonstrates exceptional accuracy, while the measurement algorithm offers superior efficiency when compared to the conventional caliper-based method utilized with spectral-domain OCT devices—a time-consuming procedure for ophthalmologists. These advancements promise to expedite the diagnostic process and facilitate to rapidly recognize the stage 2 MHs from stage 3 or 4 cases.

As an interdisciplinary area, Data Science draws scientific inquiry from a broad range of subject areas such as statistics, mathematics, computer science, machine learning, optimisation, signal processing, information retrieval, databases, cloud computing, computer vision, natural language processing, and so forth. Data Science aims to deliver valuable insights from data, and to meet the challenges of processing very large datasets, that is, Big Data, with new data continuously generated from various channels, such as smart devices, web, mobile and social media. Data volumes of applications in the fields of sciences and engineering, finance, media, online information resources, and so forth, are expected to double every two years over the next decade and beyond. The importance of data intensive systems has been raising and will continue to be one of the foremost fields of research. This brings up many research issues concerning capturing and accessing data effectively and fast, processing it whilst still achieving high performance and throughput, and storing it efficiently for future use. As such, data intensive systems pose many challenges in exploiting parallelism of current and upcoming computer architectures. This special issue focuses on recent advances in Data Science (e.g., Knowledge Discovery, Data Mining, Machine Learning, Big Data Analytics, Deep Learning, etc.) and data systems, and innovative real-world applications of such technologies to deliver effective and efficient solutions for current and future challenges. This special issue has attracted more than 20 submissions and 6 manuscripts were selected based on review reports. Each paper was reviewed by at least two reviewers and went through at least two rounds of reviews. The contributions of these papers are summarized below. The first contribution by Li et al. reports a novel weighted probabilistic frequent itemset mining algorithm in uncertain databases (i.e., w-PFI), which is implemented by an efficient candidate generation and validation paradigm similar to the working principle of Apriori. This work additionally presents a new probability model to support w-PFI candidate, and three pruning techniques to effectively remove the unpromising candidates immediately to improve system efficiency. The experimental results show that the proposed algorithm w-PFI yields the best performance amongst the referenced competitors in terms of running time and scalability. The second paper by Sadhukham and Palit presents a novel neighbourhood-based multi-label classifier based on the principles of reverse k-nearest neighbourhood. That is, the neighbourhood was estimated using the reverse k-nearest neighbourhood. This adaptive neighbourhood estimation with the support of implicit handling of the local imbalance works particularly well for multiple-label datasets with imbalanced labels. The proposed approach improves the efficacy of the compared methods based on the experimentation as evidenced by its competitive performance. The third publication by Tsinaslanidis and Guijarro considers chart pattern recognition for trading purposes. In particular, this work proposes the design of a trading system using generic pattern recognition technique which takes proven generic profitable patterns based on historical data as system inputs rather than restricting the search to specific technical patterns. The effectiveness of the proposed system was validated and evaluated by applying the approach to 560 NYSE stocks with generally promising results demonstrated. The article produced by Hu et al. documents an adaptive network with a stacked hourglass network and SSD for video pose estimation especially for videos with joint occlusion. The proposed network is supported by the optimisation of time series motion data using an outlier detection and a Kalman filter. The work was evaluated by applying the proposed adaptive network on two well-known benchmark human pose estimation datasets. The results show higher accuracy and good practicality. The next article by Naik et al. proposes a cognizant honeypot for active fingerprinting attack detection using dynamic fuzzy rule interpolation. This project firstly actively collected data using simulated attacks on honeypots and extracted the most influential attributes from the collected data as the signatures of active fingerprinting attacks. Then, the selected attributes were utilized to devise the dynamic fuzzy rule interpolation system and subsequently to implement the cognizant honeypot. The proposed system is featured by its dynamic rule base for more accurate and efficient detection. The final contribution by Gao et al. reports a hand gesture recognition approach using multimodal data fusion and a multiscale parallel convolutional neural network for human robot interaction. Ten hand gestures were considered in this project and the multiscale parallel convolutional neural network was trained using a dataset generated by this project. The proposed method was implemented on a seven-degree-of-freedom bionic manipulator and promising results were demonstrated based on the experiments using this manipulator. We would like to express our sincere thanks to Dr. Jon G. Hall (Editor-in-Chief of the Wiley-Blackwell Journal Expert Systems: The Journal of Knowledge Engineering) for providing the opportunity to edit this special issue. Additional thanks to the editorial staff for their excellent support. Finally, the guest editors would also like to thank all the referees for their thorough and constructive comments. The authors declare no conflicts of interest.

Mammograms are always used to detect signs of breast cancer. Texture-analysis techniques were applied to determine imaging biomarkers consisting of mean, contrast, correlation, energy and homogeneity features of parametric maps, and they can be utilized to retravel specific features of symptoms from medical images produced by X-ray, magnetic resonance imaging, computed tomography, and so forth. Gray level run-length matrix is the one of the texture extraction methods which has been successfully used to facilitate medical image analysis. However, it is computation-intensive method. We implemented it on GPU to accelerating extraction process for mammograms. The proposed method achieves significant speedup over CPU-based implementation.

Recently, Deep Learning has been applied to many medical domains, such as medical image analysis, bioinformatic, biochemistry, drug design, and so forth, to improve the performance that is superior to traditional computational approaches; especially in computer-aided drug design. Many AI-driven drug discovery startups have utilized deep learning methodology to achieve the significant improvement of searching candidate compounds, predicting functions, and so forth. Therefore, using AI to facilitate drug design is the trend in the coming future. In this study, a comprehensive review of the current state-of-the-art in Computer-Aided Drug Design using deep learning methods is presented. Meanwhile, the challenges and potential of these methods are also highlighted.

Recently, deep learning technologies have been utilized in many scientific domains successfully. Convolution neural networks are common used in image understanding problems. However, to train a convolution neural network model with huge amount of images is time-consuming task. Most of deep learning frameworks, such as Caffe, TensorFlow, Torch, Keras, MxNet, and so forth, support GPU to train model fast; especially executing these models on multiple GPUs. In this work, we present the comparison of computation performance of AlexNet among different GPU servers and hyperparameters. The results shows that GPU servers with high bandwidth rate, NVLINK, can achieve better performance than others.

Due to the rapid increase in biological data dimension and acquisition rate, the traditional analysis methods are unable to achieve acceptable accuracy. Recently, Deep learning technologies have shown outstanding results in many domains; especially in pattern recognition in the field of bioinformatics. In this paper, we provide background of what deep learning and its frameworks. In addition, we review the state-of-the-art algorithms based on GPU to presenting the usage of them to guide computational biologists to know how to leverage deep learning to improve their methods.

A phylogenetic tree is used to present the evolutionary relationships among the interesting biological species based on the similarities in their genetic sequences. The UPGMA is one of the popular algorithms to construct a phylogenetic tree according to the distance matrix created by the pairwise distances among taxa. To solve the performance issue of the UPGMA, the implementation of the UPGMA method on a single GPU has been proposed. However, it is not capable of handling the large taxa set. This work describes a novel parallel UPGMA approach on multiple GPUs that is able to build a tree from extremely large datasets. The experimental results show that the proposed approach with 4 NVIDIA GTX 980 achieves an approximately × fold speedup over the implementation of UPGMA on CPU and GPU, respectively.

Haze is an atmospheric phenomenon that fogs the visibility of the scenes. Removing the haze has been an important issue in image processing technologies. Many image dehazing technologies with evolution algorithms have been proposed to remove the fog in the image. However, these algorithms usually are compute-intensive. In this paper, we propose a parallel hybrid evolution algorithm based on GPU to enhance the computational performance. In traditional evolution algorithms, the calculation of fitness function occupies the most of the computation time. In the proposed method, we implement this part on GPU by using CUDA framework to reduce the computational load. The experiment results show that the proposed method can remove the haze efficiently and successfully.

Summary Wafer processing technology has been improving rapidly. Moore's law has been exceeded as the number of transistors in a dense integrated circuit, now increases threefold or more, approximately every year. The integrated circuit has gone from very large scale to giga large scale. The extraction of subcircuits has therefore become computation‐intensive. In this paper, we propose an efficient bit‐parallel subcircuit extraction algorithm using graphic processing units. We conducted experimental trials and demonstrated that the proposed algorithm can achieve high throughput, suggesting practical applications in the extraction of subcircuits. Copyright © 2015 John Wiley & Sons, Ltd.

Summary Detection of white matter changes in brain tissue using magnetic resonance imaging has been an increasingly active and challenging research area in computational neuroscience. A genetic algorithm based on a fuzzy c‐mean clustering method (GAFCM) was applied to simulated images to separate foreground spot signal information from the background, and the results were compared. The strength of this algorithm was tested by evaluating the segmentation matching factor, coefficient of determination, concordance correlation, and gene expression values. The experimental results demonstrated that the segmentation ability of GAFCM was better than that of fuzzy c‐means and K‐means algorithms. However, GAFCM is computationally expensive. This study presents a new GPU‐based parallel GAFCM algorithm to improve the performance of GAFCM. The experimental results show that computational performance can be increased by a factor of approximately 20 over the CPU‐based GAFCM algorithm while maintaining the quality of the processed images. Thus, the proposed GPU‐based parallel GAFCM algorithm can achieve the same results and significantly decrease processing time. Copyright © 2015 John Wiley & Sons, Ltd.

Embedded multicore computing and applications

With the rapid growth of biological technology, large amount of biological data can be produced in few days or months. Many of common-used tools become computation-consuming in analyzing such big biological data. Cloud computing has emerged to provide the huge amount of computing power and play important role in development of bioinformatics tools. We propose a cloud computing framework that is able to easily deploy the bioinformatics tools on cloud virtualization platform based on Hadoop. This framework can work on the public cloud platform vendor such as Amazon EC2 and also private cloud platform. All the tools performed by cloud computing framework as Bioinformatics as a Services are available at http://bioinfo.cs.pu.edu.tw/CBBTS. The tools deployed on cloud platform by the proposed framework are tested in Providence University cloud platform and provided good proportional acceleration when scaled out onto many computational units. In the big biological data era, cloud computing based solutions are important role to develop bioinformatics services over internet. In the work, the proposed framework is able to simply deploy several well-known bioinformatics tools on cloud virtualization platform and as web services. This framework can work on the public cloud platform vendor such as Amazon EC2 and also private cloud platform.

GPU computing has gradually become the mainstream to do high-speed computing fields, such as the meteorology, image and video processing, fluid dynamics simulation, seismic analysis, and etc. How to efficiently port an existing program on CPU to its CUDA program on GPU is an important issue. From the previous works, the porting approach can be generalized and classified into two categories: Rewrite Parallel Algorithm (abbreviate to RPA) and Modify Original Library (abbreviate to MOL). For the RPA, the programmers need to understand the original sequential or parallel algorithm on CPU absolutely and then write the CUDA program on GPU directly. For the MOL, the programmers need to analyze (profile) the existing program on CPU at first to find the most spend time libraries (or functions), then they are modified greatly (rewritten in general) to become CUDA programs (kernel functions). There are several disadvantages for the RPA and MOL, especially for the porting time and executing results. Hence, in this paper, a new approach, called innovative systematic contract (abbreviate to ISC), is proposed to allow programmers to port an existing CPU program to its CUDA program by modifying the libraries lightly. The program, BLASTN v2.2.27, was ported into a CUDA version, called CUDA-BLASTN v1, by the ISC. From the experimental results, by comparing with BLASTN v2.2.27, CUDA-BLASTN v1 achieves 5x speedup ratio and obtains almost the same executing results.

Summary Constructing phylogenetic trees is of priority concern in computational biology, especially for developing biological taxonomies. As a conventional means of constructing phylogenetic trees, unweighted pair group method with arithmetic (UPGMA) is also an extensively adopted heuristic algorithm for constructing ultrametric trees (UT). Although the UT constructed by UPGMA is often not a true tree unless the molecular clock assumption holds, UT is still useful for the clocklike data. Moreover, UT has been successfully adopted in other problems, including orthologous‐domain classification and multiple sequence alignment. However, previous implementations of the UPGMA method have a limited ability to handle large taxa sets efficiently. This work describes a novel graphics processing unit (GPU)‐UPGMA approach, capable of providing rapid construction of extremely large datasets for biologists. Experimental results indicate that the proposed GPU‐UPGMA approach achieves an approximately 95× speedup ratio on NVIDIA Tesla C2050 GPU over the implementation with 2.13 GHz CPU. The developed techniques in GPU‐UPGMA also can be applied to solve the classification problem for large data set with more than tens of thousands items in the future.Copyright © 2014 John Wiley & Sons, Ltd.

In recent years, botnet is one of the major threats to network security. Many approaches have been proposed to detect botnets by comparing bot features. Usually, these approaches adopt traffic reduction strategy as first step to reduce the flow to following strategies by filtering packets. With the rapid development of network hardware and software the network speed has reached to multi-gigabit. However, analyzing header and payload of every packet consumes huge amount of computational resources and is not suitable for many realistic situations. Although signature-based solutions are accurate, it is not possible to detect bot variants in real-time. In this study, we proposed a GPU-based botnet detection approach. The experimental results show that the network traffic reduction stage on GPU can achieve about 8x times over CPU based botnet detection tool. The proposed algorithm can used to improve the performance of botnet detection tools efficiently.

In the 21stcentury, antibiotic resistance has become a crucial and growing phenomenon in contemporary medicine. Multidrug resistant leads that antibiotics cannot be used to treat infections. In this paper, we propose a novel and efficient method to identify drug resistance genes from raw reads produced by next generation sequencing technology for metagenomes. The experimental results show that the proposed method is able to identify the resistance genes of Acinetobacter baumannii, TYTH-1.

Compound comparison is an important task for computational chemistry. By the comparison reulsts, potential inhibitors can be found and then used for the following experiments. The time complexity of a pairwise compound comparison is O(n2), where n is the maximal length of compounds. In general, the compound length is small, and the cost of computation time is short. However, more and more compounds have been synthesized and extracted now, even more than ten of millions. Therefore, it still will be time-consuming when comparing with a large amount of compounds (multiple compound comparisons). In this paper, we propose a parallel algorithm for multiple compound comparisons on multi-GPUs. Four load-balancing strategies were considered in the proposed algorithm in order to accelerate the computation speed among thread blocks on GPUs. The proposed algorithm was implemented by C+OpenMP+CUDA, and achieved more than 50 times speedup by comparing with its CPU version under the experiemtal results.

Nowadays, with the explosive growth of the network technologies many new applications and services have been developed on Internet. World Wide Web can provide these services provided without the limitation of time and location. Obviously, the number of user is dramatically increasing from amount of the visitations of web pages. In our previous work, we proposed an algorithm to discover more significant information from visited web pages to provide this information to web designers or policy makers to adjust the presentation of their Web contents. However, this algorithm is time-consuming approach due to it needs to scan the whole database many times. Therefore, we propose a GPGPU-based Preference Utility algorithm to enhance the performance by GPGPU parallel model. The proposed algorithm is developed on NVIDIA CUDA architecture. The experimental results show that the proposed method can achieve about 7x times over CPU-based method. The proposed algorithm can used to mine the information from web log data efficiently.

Cloud computing, an emerging Internet-based development provides various platform and software services has become a significant issue. Many bioinformatics tools developed based on the Internet for biologists without reconstructing the whole software in the local system. Therefore, bioinformatics as a service is a new significant demand of cloud computing that integrates the bioinformatics tools to cloud platform to provide more efficient bio-computing services. In this paper, we propose a virus analysis service on the cloud. Norovirus is used as case study for this service. The result demonstrates that the proposed cloud service is able to be easily used to analyze virus data by using remote computing resources.

Currently, cloud computing has been applied to share computing resources to achieve coherence and economies of scale similar to a utility over a network. Hadoop is an widely-used open-source cloud computing environment that implements the Google MapReduce framework. Many bioinformatics tools have been developed to provide cloud services by using Hadoop. This paper proposes approaches in providing a pairwise 3D protein structure alignment; our web service takes advantage of the MapReduce paradigm as means of management and parallelizing tools under massive number of protein pairs examined under the experiment. It shows that our previously proposed sequential combinatorial algorithms are well parallelized under the map/reduce platform. These methods are tested on the real-world data obtained in from the RCSB PDB data set; the computation efficiency can be effectively improved proportional to the number of processors being used.

The Smith-Waterman algorithm is the most widely used algorithm to analyze the similarity between protein and DNA sequences and suitable for the database search due to its high sensitivity. However, Smith-Waterman still is a very time-consuming method. CUDA programming can efficiently improve the computations by using the computing power of the massive computing hardware as GPUs. In this paper, we proposed an efficient frequency based filter method instead of just speed up the Smith-Waterman comparison but waste computing resource to deal with those unnecessary comparisons. We implemented the Smith-Waterman algorithm by introduction of the techniques from earlier researches and add in our real-time filter method on Graphic Processing Units to filter unnecessary comparisons. We also design a user friendly interface to provide the service in the potential clouding computing environment. In our research we choose two data sets, H1N1 VH protein database and Human protein database then compare CUDA-SW and CUDA-SW with filter, we called CUDA-SWf we can obtain up to 41% performance improve from reduce unnecessary sequence alignments.

Recently, many new next-generation sequencing techniques have been proposed. These techniques can produce lot of short reads rapidly. Hence, a number of tools have been developed to map these short reads to the genome. However, with more and more reads sequenced and the length of reads increases, these tools require high memory usage and huge computational cost and are also impractical for utilization. As the GPU has become increasingly more powerful and ubiquitous, many scientific applications have been implemented to enhance the computational performance on GPU platform. In this paper, we proposed a method, CUDA-FRESCO, to map the short reads to the genome by using CUDA on GPU platform. The experimental results present that CUDA-FRESCO can achieve dramatic speed up than other tools. CUDA-FRESCO can be alternative tool for biologists to map the short reads fast.

With the rapid growth of network technologies, many new web services have been developed to provide various applications and computing functions. These services rely deeply on the internet. Therefore, packet classification is an important issue of network security that typically adopts a flexible packet filtering system to classify each processed packet. Traditional packet classification requires hung computing time to process large amount of internet packets. Hence, we propose a GPGPU-based parallel packet classification method to decrease the computational cost. We also evaluate the performance of the proposed method with implementation on various memory architectures of CUDA device. The experiment results demonstrate that the proposed method can achieve significant speed up over the sequential packet classification algorithms on single CPU.