• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2002.06a
• /
• pp.148-155
• /
• 2002

Multithreaded models improve the efficiency of parallel systems by combining inner parallelism, asynchronous data availability and the locality of von Neumann model. This model executes thread code which is generated by compiler and of which quality is given by the method of generation. But multithreaded models have the demerit that execution model is restricted to a specific platform. On the contrary, Java has the platform independency, so if we can translate from threads code to Java bytecode, we can use the advantages of multithreaded models in many platforms. Java executes Java bytecode which is intermediate language format for Java virtual machine. Java bytecode plays a role of an intermediate language in translator and Java virtual machine work as back-end in translator. But, Java bytecode which is translated from multithreaded models have the demerit that it is not secure. This paper, multhithread code whose feature of platform independent can execute in java virtual machine. We design and implement translator which translate from thread code of multithreaded code to Java bytecode and which check secure problems from Java bytecode.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.127-148
• /
• 2020

The data center is a physical environment facility for accommodating computer systems and related components, and is an essential foundation technology for next-generation core industries such as big data, smart factories, wearables, and smart homes. In particular, with the growth of cloud computing, the proportional expansion of the data center infrastructure is inevitable. Monitoring the health of these data center facilities is a way to maintain and manage the system and prevent failure. If a failure occurs in some elements of the facility, it may affect not only the relevant equipment but also other connected equipment, and may cause enormous damage. In particular, IT facilities are irregular due to interdependence and it is difficult to know the cause. In the previous study predicting failure in data center, failure was predicted by looking at a single server as a single state without assuming that the devices were mixed. Therefore, in this study, data center failures were classified into failures occurring inside the server (Outage A) and failures occurring outside the server (Outage B), and focused on analyzing complex failures occurring within the server. Server external failures include power, cooling, user errors, etc. Since such failures can be prevented in the early stages of data center facility construction, various solutions are being developed. On the other hand, the cause of the failure occurring in the server is difficult to determine, and adequate prevention has not yet been achieved. In particular, this is the reason why server failures do not occur singularly, cause other server failures, or receive something that causes failures from other servers. In other words, while the existing studies assumed that it was a single server that did not affect the servers and analyzed the failure, in this study, the failure occurred on the assumption that it had an effect between servers. In order to define the complex failure situation in the data center, failure history data for each equipment existing in the data center was used. There are four major failures considered in this study: Network Node Down, Server Down, Windows Activation Services Down, and Database Management System Service Down. The failures that occur for each device are sorted in chronological order, and when a failure occurs in a specific equipment, if a failure occurs in a specific equipment within 5 minutes from the time of occurrence, it is defined that the failure occurs simultaneously. After configuring the sequence for the devices that have failed at the same time, 5 devices that frequently occur simultaneously within the configured sequence were selected, and the case where the selected devices failed at the same time was confirmed through visualization. Since the server resource information collected for failure analysis is in units of time series and has flow, we used Long Short-term Memory (LSTM), a deep learning algorithm that can predict the next state through the previous state. In addition, unlike a single server, the Hierarchical Attention Network deep learning model structure was used in consideration of the fact that the level of multiple failures for each server is different. This algorithm is a method of increasing the prediction accuracy by giving weight to the server as the impact on the failure increases. The study began with defining the type of failure and selecting the analysis target. In the first experiment, the same collected data was assumed as a single server state and a multiple server state, and compared and analyzed. The second experiment improved the prediction accuracy in the case of a complex server by optimizing each server threshold. In the first experiment, which assumed each of a single server and multiple servers, in the case of a single server, it was predicted that three of the five servers did not have a failure even though the actual failure occurred. However, assuming multiple servers, all five servers were predicted to have failed. As a result of the experiment, the hypothesis that there is an effect between servers is proven. As a result of this study, it was confirmed that the prediction performance was superior when the multiple servers were assumed than when the single server was assumed. In particular, applying the Hierarchical Attention Network algorithm, assuming that the effects of each server will be different, played a role in improving the analysis effect. In addition, by applying a different threshold for each server, the prediction accuracy could be improved. This study showed that failures that are difficult to determine the cause can be predicted through historical data, and a model that can predict failures occurring in servers in data centers is presented. It is expected that the occurrence of disability can be prevented in advance using the results of this study.

• Asian-Australasian Journal of Animal Sciences
• /
• v.30 no.7
• /
• pp.967-972
• /
• 2017

Objective: Two experiments were conducted to determine the effects of forage-to-concentrate (F:C) ratio on the nutrient digestibility and enteric methane ($CH_4$) emission in growing goats and Sika deer. Methods: Three male growing goats (body weight $[BW]=19.0{\pm}0.7kg$) and three male growing deer ($BW=19.3{\pm}1.2kg$) were respectively allotted to a $3{\times}3$ Latin square design with an adaptation period of 7 d and a data collection period of 3 d. Respiration-metabolism chambers were used for measuring the enteric $CH_4$ emission. Treatments of low (25:75), moderate (50:50), and high (73:27) F:C ratios were given to both goats and Sika deer. Results: Dry matter (DM) and organic matter (OM) digestibility decreased linearly with increasing F:C ratio in both goats and Sika deer. In both goats and Sika deer, the $CH_4$ emissions expressed as g/d, g/kg $BW^{0.75}$, % of gross energy intake, g/kg DM intake (DMI), and g/kg OM intake (OMI) decreased linearly as the F:C ratio increased, however, the $CH_4$ emissions expressed as g/kg digested DMI and OMI were not affected by the F:C ratio. Eight equations were derived for predicting the enteric $CH_4$ emission from goats and Sika deer. For goat, equation 1 was found to be of the highest accuracy: $CH_4(g/d)=3.36+4.71{\times}DMI(kg/d)-0.0036{\times}neutral$ detergent fiber concentrate (NDFC,g/kg)+$0.01563{\times}dry$ matter digestibility (DMD,g/kg)-$0.0108{\times}neutral$ detergent fiber digestibility (NDFD, g/kg). For Sika deer, equation 5 was found to be of the highest accuracy: $CH_4(g/d)=66.3+27.7{\times}DMI(kg/d)-5.91{\times}NDFC(g/kg)-7.11{\times}DMD(g/kg)+0.0809{\times}NDFD(g/kg)$. Conclusion: Digested nutrient intake could be considered when determining the $CH_4$ generation factor in goats and Sika deer. Finally, the enteric $CH_4$ prediction model for goats and Sika deer were estimated.

• Journal of Environmental Science International
• /
• v.15 no.8
• /
• pp.775-784
• /
• 2006

In this study, CN value was estimated by using detailed soil map and land cover characteristic against upper basin of Kumho watermark located on the upper basin of Kumho river and the hydrologic morphological characteristic factors were extracted from the basin by using the DEM document. Also the runoff analysis was conducted by the WMS model in order to study how the assumed CN value affects the runoff characteristic. First of all, as a result of studying the soil type in this study area, mostly D type soil was Identified by the application of the 1987 classification criteria. However, by that in 1995, B type soil and C type soil were distributed more widely in that area. When CN value was classified by the 1995 classification criteria, it was estimated lower than in 1987, as a result of comparing the estimated CNs by those standars. Also it was assumed that CN value was underestimated when the plan for Geum-ho river maintenance was drawn up. As a result of the analysis of runoff characteristic, the pattern of generation of the classification criteria of soil groups appeared to be similar, but in the case of the application of the classification criteria in 1995, the peak rate of runoff was found to be smaller on the whole than in the case of the application of the classification criteria in 1987. Also when the statistical data such as the prediction errors, the mean squared errors, the coefficient of determination and other data emerging from the analysis, was looked over in total, it seemed appropriate to apply the 1995 classification criteria when hydrological soil classification group was applied. As the result of this study, however, the difference of the result of the statistical dat was somewhat small. In future study, it is necessary to follow up evidence about soil application On many more watersheds and in heavy rain.

• Journal of IKEEE
• /
• v.23 no.2
• /
• pp.438-447
• /
• 2019

As the IoT-based hyper-connected society grows, public-key cryptosystem such as RSA is frequently used for encryption, authentication, and digital signature. Public-key cryptosystem use very large (safe) prime numbers to ensure security against malicious attacks. Even though the performance of the device has greatly improved, the generation of a large (safe)prime is time-consuming or memory-intensive. In this paper, we propose ET-MR and ET-MR-MR using Euler sieve so it runs faster while using less memory. We present a running time prediction model by probabilistic analysis and compare time and memory of our method with conventional methods. Experimental results show that the difference between the expected running time and the measured running time is less than 4%. In addition, the fastest running time of ET-MR is 36% faster than that of TD-MR, 8.5% faster than that of DT-MR and the fastest running time of ET-MR-MR is 65.3% faster than that of TD-MR-MR and similar to that of DT-MR-MR. When k=12,381, the memory usage of ET-MR is 2.7 times more than that of DT-MR but 98.5% less than that of TD-MR and when k=65,536, the memory usage of ET-MR-MR is 98.48% less than that of TD-MR-MR and 92.8% less than that of DT-MR-MR.

• Journal of the Korean GEO-environmental Society
• /
• v.12 no.9
• /
• pp.55-61
• /
• 2011

This study investigate the use of ultraviolet(UV) light with hydrogen peroxide($H_2O_2$) for Methyl Tert Butyl Ether(MTBE) degradation in photolysis reactor. The process in general demands the generation of OH radicals in solution at the presence of UV light. These radicals can then attack the MTBE molecule and it is finally destroyed or converted into a simple harmless compound. The MTBE removal by photolysis were mathematically described as the independent variables such as irradiation intensity, initial concentration of MTBE and $H_2O_2$/MTBE ratio, and these were modeled by the use of response surface methodology(RSM). These experiments were carried out as a Box-Behnken Design(BBD) consisting of 15 experiments. Regression analysis term of Analysis of Variance(ANOVA) shows significantly p-value(p<0.05) and high coefficients for determination values($R^2$=94.60%) that allow satisfactory prediction of second-order regression model. And Canonical analysis yields the stationery point for response, with the estimate ridge of maximum responses and optimal conditions for Y(MTBE removal efficiency, %) are $x_1$=25.75 W of irradiation intensity, $x_2$=7.69 mg/L of MTBE concentration and $x_3$=11.04 of $H_2O_2$/MTBE molecular ratio, respectively. This study clearly shows that RSM is available tool for optimizing the operating conditions to maximize MTBE removal.

• Nuclear Engineering and Technology
• /
• v.27 no.6
• /
• pp.811-824
• /
• 1995

The CANDU element bowing is attributed to actions of both the thermally induced bending moments and the bending moment due to hydraulic drag and mechanical loads, where the bowing is defined as the lateral deflection of an element from the axial centerline. This paper consider only the thermally-induced bending moments which are generated both within the sheath and the fuel and sheath by an asymmetric temperature distribution with respect to the axis of an element The generalized and explicit analytical formula for the thermally-induced bending is presented in con-sideration of 1) bending of an empty tube treated by neglecting the fuel/sheath mechanical interaction and 2) fuel/sheath interaction due to the pellet and sheath temperature variations, where in each case the temperature asymmetries in sheath are modelled to be caused by the combined effects of (i) non-uniform coolant temperature due to imperfect coolant mixing, (ii) variable sheath/coolant heat transfer coefficient, (iii) asymmetric heat generation due to neutron flux gradients across an element and so as to inclusively cover the uniform temperature distributions within the fuel and sheath with respect to the axial centerline. As the results of the sensitivity calculations of the element bowing with the variations of the parameters in the formula, it is found that the element bowing is greatly affected relatively with the variations or changes of element length, sheath inside diameter, average coolant temperature and its variation factor, pellet/sheath mechanical interaction factor, neutron flux depression factor, pellet thermal expansion coefficient, pellet/sheath heat transfer coefficient in comparison with those of other parameters such as sheath thickness, film heat transfer coefficient, sheath thermal expansion coefficient and sheath and pellet thermal conductivities.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.3
• /
• pp.250-261
• /
• 2023

In this study, the noise contributions by the duct, stator and rotor, which are the propulsor components, are evaluated to identify the flow noise source in cavitation and non-cavitation conditions on pumpjet propulsion and the noise levels in both conditions are compared. The unsteady incompressible Reynolds averaged Navier-Stokes (RANS) equation based on the homogeneous mixture assumption is applied on the suboff submarine hull and pumpjet propeller in the cavitation tunnel, and the Volume of Fluid (VOF) method and Schnerr-Sauer cavitation model are used to describe the two-phase flow. Based on the flow simulation results, the acoustic analogy formulated by Ffowcs Williams and Hawkings (FW-H) equation is applied to predict the underwater radiated noise. The noise contributions are evaluated by using the three types of impermeable integral surface on the duct, stator and rotor, and the two types of permeable integral surface surrounding the propulsor. As a result of noise prediction, the contribution by the stator is insignificant, but it affects the generation of flow noise source due to flow separation in the duct and rotor, and the noise is predominantly radiated into the upward and right where the flow separations are. Also, the noise is radiated into the thrust direction due to pressure fluctuation between suction and pressure sides on the rotor blades, and the it can be seen that the cavitation effect into the noise can be considered through the permeable integral surface.

• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.1-19
• /
• 2018

Large amount of data is now available for research and business sectors to extract knowledge from it. This data can be in the form of unstructured data such as audio, text, and image data and can be analyzed by deep learning methodology. Deep learning is now widely used for various estimation, classification, and prediction problems. Especially, fashion business adopts deep learning techniques for apparel recognition, apparel search and retrieval engine, and automatic product recommendation. The core model of these applications is the image classification using Convolutional Neural Networks (CNN). CNN is made up of neurons which learn parameters such as weights while inputs come through and reach outputs. CNN has layer structure which is best suited for image classification as it is comprised of convolutional layer for generating feature maps, pooling layer for reducing the dimensionality of feature maps, and fully-connected layer for classifying the extracted features. However, most of the classification models have been trained using online product image, which is taken under controlled situation such as apparel image itself or professional model wearing apparel. This image may not be an effective way to train the classification model considering the situation when one might want to classify street fashion image or walking image, which is taken in uncontrolled situation and involves people's movement and unexpected pose. Therefore, we propose to train the model with runway apparel image dataset which captures mobility. This will allow the classification model to be trained with far more variable data and enhance the adaptation with diverse query image. To achieve both convergence and generalization of the model, we apply Transfer Learning on our training network. As Transfer Learning in CNN is composed of pre-training and fine-tuning stages, we divide the training step into two. First, we pre-train our architecture with large-scale dataset, ImageNet dataset, which consists of 1.2 million images with 1000 categories including animals, plants, activities, materials, instrumentations, scenes, and foods. We use GoogLeNet for our main architecture as it has achieved great accuracy with efficiency in ImageNet Large Scale Visual Recognition Challenge (ILSVRC). Second, we fine-tune the network with our own runway image dataset. For the runway image dataset, we could not find any previously and publicly made dataset, so we collect the dataset from Google Image Search attaining 2426 images of 32 major fashion brands including Anna Molinari, Balenciaga, Balmain, Brioni, Burberry, Celine, Chanel, Chloe, Christian Dior, Cividini, Dolce and Gabbana, Emilio Pucci, Ermenegildo, Fendi, Giuliana Teso, Gucci, Issey Miyake, Kenzo, Leonard, Louis Vuitton, Marc Jacobs, Marni, Max Mara, Missoni, Moschino, Ralph Lauren, Roberto Cavalli, Sonia Rykiel, Stella McCartney, Valentino, Versace, and Yve Saint Laurent. We perform 10-folded experiments to consider the random generation of training data, and our proposed model has achieved accuracy of 67.2% on final test. Our research suggests several advantages over previous related studies as to our best knowledge, there haven't been any previous studies which trained the network for apparel image classification based on runway image dataset. We suggest the idea of training model with image capturing all the possible postures, which is denoted as mobility, by using our own runway apparel image dataset. Moreover, by applying Transfer Learning and using checkpoint and parameters provided by Tensorflow Slim, we could save time spent on training the classification model as taking 6 minutes per experiment to train the classifier. This model can be used in many business applications where the query image can be runway image, product image, or street fashion image. To be specific, runway query image can be used for mobile application service during fashion week to facilitate brand search, street style query image can be classified during fashion editorial task to classify and label the brand or style, and website query image can be processed by e-commerce multi-complex service providing item information or recommending similar item.

• KDI Journal of Economic Policy
• /
• v.30 no.2
• /
• pp.225-270
• /
• 2008

We analyze macroeconomic consequences of pay-as-you-go (PAYGO) public pension system with a simple overlapping generations model. Contrary to large body of existing literatures offering quantitative results based on simulation study, we take another route by adopting a highly simplified framework in search of qualitatively tractable analytical results. The main contribution of our results lies in providing a sound theoretical foundation that can be utilized in interpreting various quantitative results offered by simulation studies of large scale general equilibrium models. We present a simple overlapping generations model with a defined benefit(DB) PAYGO public pension system as a benchmark case and derive an analytical equilibrium solution utilizing graphical illustration. We also discuss the modifications of the benchmark model required to encompass a defined contribution(DC) public pension system into the basic framework. Comparative statics analysis provides three important implications; First, introduction and expansion of the PAYGO public pension, DB or DC, result in lower level of capital accumulation and higher expected rate of return on the risky asset. Second, it is shown that the progress of population aging is accompanied by lower capital stock due to decrease in both demand and supply of risky asset. Moreover, risk premium for risky asset increases(decreases) as the speed of population aging accelerates(decelerates) so that the possibility of so-called "the great meltdown" of asset market cannot be excluded although the odds are not high. Third, it is most likely that the switch from DB PAYGO to DC PAYGO would result in lower capital stock and higher expected return on the risky asset mainly due to the fact that the young generation regards DC PAYGO pension as another risky asset competing against the risky asset traded in the market. This theoretical prediction coincides with one of the firmly established propositions in empirical literature that the currently dominant form of public pension system has the tendency to crowd out private capital accumulation.