• The Korean Journal of Applied Statistics
• /
• v.27 no.2
• /
• pp.249-261
• /
• 2014

Ever increasing "Big data" can only be effectively processed by parallel computing. Parallel computing refers to a high performance computational method that achieves effectiveness by dividing a big query into smaller subtasks and aggregating results from subtasks to provide an output. However, it is well-known that parallel computing does not achieve scalability which means that performance is improved linearly by adding more computers because it requires a very careful assignment of tasks to each node and collecting results in a timely manner. Hadoop is one of the most successful platforms to attain scalability. In this paper, we propose a measurement for Hadoop optimization by utilizing a Lorenz curve which is a proxy for the inequality of hardware resources. Our proposed index takes into account the intrinsic overhead of Hadoop systems such as CPU, disk I/O and network. Therefore, it also indicates that a given Hadoop can be improved explicitly and in what capacity. Our proposed method is illustrated with experimental data and substantiated by Monte Carlo simulations.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.6
• /
• pp.1431-1438
• /
• 2014

We analyzed the articles from "Kukje Shinmun" and "Busan Ilbo", which are two local newpapers of Busan Metropolitan City. The articles cover from January 1, 2013 to December 31, 2013. Meaningful pattern inherent in 2889 articles of which the title includes "Busan" and "Traffic" and related data was analyzed. Textmining method, which is a part of datamining, was used for the social network analysis (SNA). HDFS and MapReduce (from Hadoop ecosystem), which is open-source framework based on JAVA, were used with Linux environment (Uubntu-12.04LTS) for the construction of unstructured data and the storage, process and the analysis of big data. We implemented new algorithm that shows better visualization compared with the default one from R package, by providing the color and thickness based on the weight from each node and line connecting the nodes.

• Journal of Information Processing Systems
• /
• v.20 no.3
• /
• pp.375-390
• /
• 2024

Edge computing architecture has effectively alleviated the computing pressure on cloud platforms, reduced network bandwidth consumption, and improved the quality of service for user experience; however, it has also introduced new security issues. Existing anomaly detection methods in big data scenarios with cloud-edge computing collaboration face several challenges, such as sample imbalance, difficulty in dealing with complex network traffic attacks, and difficulty in effectively training large-scale data or overly complex deep-learning network models. A lightweight deep-learning model was proposed to address these challenges. First, normalization on the user side was used to preprocess the traffic data. On the edge side, a trained Wasserstein generative adversarial network (WGAN) was used to supplement the data samples, which effectively alleviates the imbalance issue of a few types of samples while occupying a small amount of edge-computing resources. Finally, a trained lightweight deep learning network model is deployed on the edge side, and the preprocessed and expanded local data are used to fine-tune the trained model. This ensures that the data of each edge node are more consistent with the local characteristics, effectively improving the system's detection ability. In the designed lightweight deep learning network model, two sets of convolutional pooling layers of convolutional neural networks (CNN) were used to extract spatial features. The bidirectional long short-term memory network (BiLSTM) was used to collect time sequence features, and the weight of traffic features was adjusted through the attention mechanism, improving the model's ability to identify abnormal traffic features. The proposed model was experimentally demonstrated using the NSL-KDD, UNSW-NB15, and CIC-ISD2018 datasets. The accuracies of the proposed model on the three datasets were as high as 0.974, 0.925, and 0.953, respectively, showing superior accuracy to other comparative models. The proposed lightweight deep learning network model has good application prospects for anomaly traffic detection in cloud-edge collaborative computing architectures.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.5 s.37
• /
• pp.123-132
• /
• 2005

In this Paper, we present an improvement of switch node for wavelength multiplex optical network. Currently because of quick increase of internet traffic a big network capacity is demanded. Wavelength multiplex optical network Provides the data transfer of high speed and the transparent characteristic of the data. Therefore optic network configuration is the most powerful technology in the future. It will be able to control the massive traffic from the optical network in order to transmit the multimedia information of very many quantify. Consequently the node where the traffic control is Possible, is demanded. The optical switch node which manages efficiently the multiple wavelength was Proposed. This switch is composed of a optical switch module for switching and a wavelength converter module for wavelength conversion. It will be able to compose the switch fabric without optical/electro or electro/optical conversion using optical MEMS(Micro Electro Mechanical Switches) module. Finally, we present the good test result regarding the operational qualify of the switch fabric and the performance of optical signal from the switch node. The proposed switch node of the optic network will be able to control the massive traffic with all optical.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.7
• /
• pp.230-238
• /
• 2014

In this paper, we propose the possible way of accurate analysis and examination of ITO sensor to discriminate whether mutual capacitance ITO sensor is defective by using mutual capacitance of data in each node which consists of electrodes inside of ITO sensor. We have analyzed the structure characteristic of mutual capacitance ITO sensor which is used as an input device for not only small size electronics like mobile phone and tablets but also big size electronics and designed the circuit to inspect ITO sensor using touch screen panel IC. Set a variable related with mutual capacitance of charge and discharge and Implement to find and analyze accurate position when defect is made through the data from each node of ITO sensor. First, we can set a yield effective range through the first experiment data of mutual capacitance ITO sensor and by using the data of each node of ITO sensor which is the result from the second experiment, we can verify accuracy and effectiveness of effective range from the first experiment as a sample which is used in this experiment.

• IEIE Transactions on Smart Processing and Computing
• /
• v.3 no.5
• /
• pp.338-344
• /
• 2014

As the role of information collection becomes increasingly important in the enormous data environment, there is growing demand for more intelligent information technologies for managing complex data. On the other hand, it is difficult to find a solution because of the data complexity and big scaled amount. Accordingly, there is a need for a special intelligent knowledge base frame that can be operated by itself flexibly. In this paper, by adopting switching function for signal transmission in the synapse of the human brain, I/E selective activation based knowledge reconfiguring mechanism is proposed for building more intelligent information management system. In particular, knowledge network design, a special knowledge node structure, Type definition, I/E gauge definition and I/E matching scheme are provided. Using these concepts, the proposed system makes the functions of activation by I/E Gauge, selection and reconfiguration. In a more efficient manner, the routing and reasoning process was performed based on the knowledge reconfiguration network. In the experiments, the process of selection by I/E matching, knowledge reconfiguration and routing & reasoning results are described.

• Journal of Industrial Technology
• /
• v.21 no.A
• /
• pp.241-249
• /
• 2001

The objective of this paper is to develop the optimal path algorithm for dynamic route guidance system in advanced traveler information system (ATIS). The travel time is forecasted in each path between network nodes. Floyd-Warshall algorithm is used to find the optimal route based on this forecasted travel time in dynamic traffic network. This algorithm is modified to apply the real traffic network that has left-turn restriction, U-turn, and P-turn. A big value is assigned to one of arcs in turn restriction and a virtual node is used to consider U-turn and P-turn for Floyd-Warshall algorithm.

• Proceedings of the KIEE Conference
• /
• 1998.07f
• /
• pp.2109-2111
• /
• 1998

This paper proposes a new lossless snubber circuit on power supply for welding machine's output rectification diode. To improve the common inverter control type power supplies' problems that energy loss and heating in the snubber circuit because the output capacity makes too big heat energy in the circuit when the output current of the inverter is rectified by the diode bridge circuit, which includes the snubber circuit. This paper suggested new snubber circuit have increased power factor and confidence of output by being regenerate thus lost energy to input node.

• Electronics and Telecommunications Trends
• /
• v.37 no.1
• /
• pp.42-52
• /
• 2022

With the advent of the big data era, the memory capacity required for computing systems is rapidly increasing, especially in High Performance Computing systems. However, the number of DRAMs that can be used in a computing node is limited by the structural limitations of the hardware (for example, CPU specifications). Memory expansion technology has attracted attention as a means of overcoming this limitation. This technology expands the memory capacity by leveraging the external memory connected to the host system through hardware interface such as PCIe and CCIX. In this paper, we present an overview and describe the development trends of the memory expansion technology. We also provide detailed descriptions and use cases of the CCIX that provides higher bandwidth and lower latency than cases of the PCIe.

• The KIPS Transactions:PartC
• /
• v.16C no.6
• /
• pp.699-706
• /
• 2009

The convergence tendency accelerates the realization of the ubiquitous healthcare (u-Healthcare) between the technology including the power generaation and IT-BT-NT of the ubiquitous computing technology. By rapidly analyzing a large amount of collected from the sensor network with processing and delivering to the medical team an u-Healthcare can provide a patient for an inappropriate regardless of the time and place. As to the existing u-Healthcare, since the sensor node all transmitted collected data by using with the Zigbee protocol the processing burden of the base node was big and there was many communication frequency of the sensor node. In this paper, the u-Healthcare system in which it can efficiently apply to mobile apparatuses it provided the transfer rate in which it is superior to the bio-signal delivery where there are the life and direct relation which by using the Bluetooth instead of the Zigbee protocol and in which it is variously used in the ubiquitous environment was designed. Moreover, by applying the EEF(Embedded Event Filtering) technique in which data in which it includes in the event defined in advance selected and it transmits with the base node, the communication frequency and were reduced. We confirmed to be the system in which it is efficient through the simulation result than the existing Electrocardiogram Measurement system.