• Journal of KIISE
• /
• v.44 no.8
• /
• pp.767-773
• /
• 2017

In a distributed system environment, such as a cloud database, the tail latency needs to be kept short to ensure uniform quality of service. In this paper, through experiments on a Cassandra database, we show that long tail latency is caused by a lack of memory space because the database cannot receive any request until free space is reclaimed by writing the buffered data to the storage device. We observed that, since the performance of the storage device determines the amount of time required for writing the buffered data, the performance degradation of Solid State Drive (SSD) due to garbage collection results in a longer tail latency. We propose a garbage collection synchronization technique, called SyncGC, that simultaneously performs garbage collection in the java virtual machine and in the garbage collection in SSD concurrently, thus hiding garbage collection overheads in the SSD. Our evaluations on real SSDs show that SyncGC reduces the tail latency of $99.9^{th}$ and, $99.9^{th}-percentile$ by 31% and 36%, respectively.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.259-259
• /
• 2019

시계열 자료들을 분석하고자 하는 경우 자료가 정상성(stationarity)을 만족하는 경우는 드물다. 특히 계절성을 제거한 자료들에서는 정량화하기 어려운 주기성이 많이 관찰된다. 즉, 어떤 특정지역에서 나타나는 현상이 다른 기상 현상에 영향을 미칠 것은 자명한 일이나 그 관련성이 선형(linearity)일 가능성은 극히 드물다. 따라서 그들 사이의 관련성이 선형성에 근거한 지표들로 정량화되어야 한다. 이러한 문제점을 해결하기 위해서 다양한 방법이 사용되며 그중에서 웨이블릿 분석을 통해 본 연구를 진행하였다. 웨이블릿 변환(wavelet transforms)은 특수한 함수의 집합으로 구성되어 기존 웨이블릿 신호의 분석을 위해 사용되는 방법이다. 이 변환은 푸리에 변환에서 변형된 방법으로 특정한 기저 함수(base function)를 이용하여 기존의 시계열 자료를 주파수로 바꾸는 변환이다. 웨이블릿 변환에서 기저 함수를 모 웨이블릿이라고 하며 이를 천이, 확대 및 축소 과정을 통해 주파수를 구성한다. 웨이블릿 분석은 모 웨이블릿을 분해하고 재결합하여 시계열 분석을 할 수 있다. 모 웨이블릿 함수에는 Haar, Daubechies, Coiflets, Symlets, Morlet, Mexican Hat, Meyer 등의 여러 가지 종류의 모 웨이블릿 함수가 있으며 모 웨이블릿이 달라지면 결과가 다르게 나타난다. 기존에는 Morlet 웨이블릿을 주로 이용하여 주파수분석에 사용하여 결과를 도출하였다. 그리고 시계열 자료는 크게 백색잡음(White Noise), 장기기억(Long Term Memory), 단기기억(Short Term Memory)으로 나뉜다. 각 시계열 자료의 종류에 따라 임의의 시계열 자료를 산정하여 그에 따른 웨이블릿 분석을 통해 모 웨이블릿의 특성을 도출하였다. 본 연구에서는 웨이블릿 분석을 통해 시계열 자료의 최적 모 웨이블릿을 결정하고자 남방진동지수(SOI), 북극진동지수(AOI)의 자료를 이용하여 웨이블릿 분석을 시도하였다. 웨이블릿 분석은 모 웨이블릿에 따라 달라지는 결과를 토대로 분석하였으며 이를 정상성과 지속성에 따라 분류된 시계열에 적용하여 최적 모 웨이블릿을 결정하고자 하였다. 본 연구에서는 임의의 시계열 자료에서 설정한 최적의 모 웨이블릿을 AOI와 SOI와 같은 실제 시계열 자료에 대입하여 분석을 진행하였다. 본 연구에서는 시계열 자료의 종류를 구분하고 자료의 특성에 따라 가장 적합한 모 웨이블릿을 구하고자 하였다.

• The Journal of Information Systems
• /
• v.31 no.3
• /
• pp.197-220
• /
• 2022

Purpose The purpose of this study is to identifies the demand pattern categorization of repair parts of Automotive After-sales Service(A/S) and proposes a demand prediction model for Auto repair parts using Long Short-Term Memory (LSTM) of artificial neural networks (ANN). The optimal parts inventory quantity prediction model is implemented by applying daily, weekly, and monthly the parts demand data to the LSTM model for the Lumpy demand which is irregularly in a specific period among repair parts of the Automotive A/S service. Design/methodology/approach This study classified the four demand pattern categorization with 2 years demand time-series data of repair parts according to the Average demand interval(ADI) and coefficient of variation (CV²) of demand size. Of the 16,295 parts in the A/S service shop studied, 96.5% had a Lumpy demand pattern that large quantities occurred at a specific period. lumpy demand pattern's repair parts in the last three years is predicted by applying them to the LSTM for daily, weekly, and monthly time-series data. as the model prediction performance evaluation index, MAPE, RMSE, and RMSLE that can measure the error between the predicted value and the actual value were used. Findings As a result of this study, Daily time-series data were excellently predicted as indicators with the lowest MAPE, RMSE, and RMSLE values, followed by Weekly and Monthly time-series data. This is due to the decrease in training data for Weekly and Monthly. even if the demand period is extended to get the training data, the prediction performance is still low due to the discontinuation of current vehicle models and the use of alternative parts that they are contributed to no more demand. Therefore, sufficient training data is important, but the selection of the prediction demand period is also a critical factor.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.216-218
• /
• 2022

Natural disasters caused by abnormal climates are continuously increasing, and the types of natural disasters that cause the most damage are flood damage caused by heavy rains and typhoons. Therefore, in order to reduce flood damage, this paper proposes a system that can predict the water level, a major parameter of flood, in real time using LSTM and GRU. The input data used for flood prediction are upstream and downstream water levels, temperature, humidity, and precipitation, and real-time prediction is performed through the pre-trained LSTM-GRU model. The input data uses data from the past 20 hours to predict the water level for the next 3 hours. Through the system proposed in this paper, if the risk determination function can be added and an evacuation order can be issued to the people exposed to the flood, it is thought that a lot of damage caused by the flood can be reduced.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.27 no.11
• /
• pp.1012-1018
• /
• 2016

We perform a electromagnetic analysis method for indirect effects of a high-conductive structure such as an aircraft exposed by lightning, by using the finite-difference time-domain(FDTD) method. The lightning waveform used to analyze indirect effects has low frequency spectrum and high-conductive materials such as aluminum and carbon fiber composite materials have very short skin depths, and thus, it requires large memory and long computation time using conventional three dimensional FDTD analysis method. We develop an efficient electromagnetic analysis method suitable for lightning and high-conductive structures. The developed analysis method is based on two dimensional FDTD and impedance network boundary condition(INBC) algorithms and we investigate the indirect effects on the structures exposed to lightning.

• Smart Media Journal
• /
• v.6 no.3
• /
• pp.41-48
• /
• 2017

Event detection using social media has been widespread since social network services have been an active communication channel for connecting with others, diffusing news message. Especially, the real-time characteristic of social media has created the opportunity for supporting for real-time applications/systems. Social network such as Twitter is the potential data source to explore useful information by mining messages posted by the user community. This paper proposed a novel system for temporal event detection by analyzing social data. As a result, this information can be used by first responders, decision makers, or news agents to gain insight of the situation. The proposed approach takes advantages of deep learning methods that play core techniques on the main tasks including informative data identifying from a noisy environment and temporal event detection. The former is the responsibility of Convolutional Neural Network model trained from labeled Twitter data. The latter is for event detection supported by Recurrent Neural Network module. We demonstrated our approach and experimental results on the case study of earthquake situations. Our system is more adaptive than other systems used traditional methods since deep learning enables to extract the features of data without spending lots of time constructing feature by hand. This benefit makes our approach adaptive to extend to a new context of practice. Moreover, the proposed system promised to respond to acceptable delay within several minutes that will helpful mean for supporting news channel agents or belief plan in case of disaster events.

• Journal of Convergence for Information Technology
• /
• v.11 no.3
• /
• pp.7-13
• /
• 2021

This study develops an artificial intelligence prediction system for Fine particulate Matter(PM2.5) based on the deep learning algorithm GAN model. The experimental data are closely related to the changes in temperature, humidity, wind speed, and atmospheric pressure generated by the time series axis and the concentration of air pollutants such as SO2, CO, O3, NO2, and PM10. Due to the characteristics of the data, since the concentration at the current time is affected by the concentration at the previous time, a predictive model for recursive supervised learning was applied. For comparative analysis of the accuracy of the existing models, CNN and LSTM, the difference between observation value and prediction value was analyzed and visualized. As a result of performance analysis, it was confirmed that the proposed GAN improved to 15.8%, 10.9%, and 5.5% in the evaluation items RMSE, MAPE, and IOA compared to LSTM, respectively.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.358-358
• /
• 2023

본 연구에서는 그동안 수자원분야 강우유출 해석분야에 활용되었던 대표적인 머신러닝&딥러닝(ML&DL) 모델을 활용하여 모델의 하이퍼파라미터 튜닝뿐만 아니라 모델의 특성을 고려한 기상 및 수문데이터의 조합과 전처리(lag-time, 이동평균 등)를 통하여 데이터 특성과 ML&DL모델의 조합시나리오에 따른 일 유입량 예측성능을 비교 검토하는 연구를 수행하였다. 이를 위해 소양강댐 유역을 대상으로 1974년에서 2021년까지 축적된 기상 및 수문데이터를 활용하여 1) 강우, 2) 유입량, 3) 기상자료를 주요 영향변수(독립변수)로 고려하고, 이에 a) 지체시간(lag-time), b) 이동평균, c) 유입량의 성분분리조건을 적용하여 총 36가지 시나리오 조합을 ML&DL의 입력자료로 활용하였다. ML&DL 모델은 1) Linear Regression(LR), 2) Lasso, 3) Ridge, 4) SVR(Support Vector Regression), 5) Random Forest(RF), 6) LGBM(Light Gradient Boosting Model), 7) XGBoost의 7가지 ML방법과 8) LSTM(Long Short-Term Memory models), 9) TCN(Temporal Convolutional Network), 10) LSTM-TCN의 3가지 DL 방법, 총 10가지 ML&DL모델을 비교 검토하여 일유입량 예측을 위한 가장 적합한 데이터 조합 특성과 ML&DL모델을 성능평가와 함께 제시하였다. 학습된 모형의 유입량 예측 결과를 비교·분석한 결과, 소양강댐 유역에서는 딥러닝 중에서는 TCN모형이 가장 우수한 성능을 보였고(TCN>TCN-LSTM>LSTM), 트리기반 머신러닝중에서는 Random Forest와 LGBM이 우수한 성능을 보였으며(RF, LGBM>XGB), SVR도 LGBM수준의 우수한 성능을 나타내었다. LR, Lasso, Ridge 세가지 Regression모형은 상대적으로 낮은 성능을 보였다. 또한 소양강댐 댐유입량 예측에 대하여 강우, 유입량, 기상계열을 36가지로 조합한 결과, 입력자료에 lag-time이 적용된 강우계열의 조합 분석에서 세가지 Regression모델을 제외한 모든 모형에서 NSE(Nash-Sutcliffe Efficiency) 0.8이상(최대 0.867)의 성능을 보였으며, lag-time이 적용된 강우와 유입량계열을 조합했을 경우 NSE 0.85이상(최대 0.901)의 더 우수한 성능을 보였다.

• Structural Engineering and Mechanics
• /
• v.85 no.4
• /
• pp.469-484
• /
• 2023

A deep recursive bidirectional Cuda Deep Neural Network Long Short Term Memory (Bi-CuDNNLSTM) layer is recruited in this paper to predict the entire force time histories, and the corresponding hysteresis and backbone curves of reinforced concrete (RC) bridge piers using experimental fast and slow cyclic tests. The proposed stacked Bi-CuDNNLSTM layers involve multiple uncertain input variables, including horizontal actuator displacements, vertical actuators axial loads, the effective height of the bridge pier, the moment of inertia, and mass. The functional application programming interface in the Keras Python library is utilized to develop a deep learning model considering all the above various input attributes. To have a robust and reliable prediction, the dataset for both the fast and slow cyclic tests is split into three mutually exclusive subsets of training, validation, and testing (unseen). The whole datasets include 17 RC bridge piers tested experimentally ten for fast and seven for slow cyclic tests. The results bring to light that the mean absolute error, as a loss function, is monotonically decreased to zero for both the training and validation datasets after 5000 epochs, and a high level of correlation is observed between the predicted and the experimentally measured values of the force time histories for all the datasets, more than 90%. It can be concluded that the maximum mean of the normalized error, obtained through Box-Whisker plot and Gaussian distribution of normalized error, associated with unseen data is about 10% and 3% for the fast and slow cyclic tests, respectively. In recapitulation, it brings to an end that the stacked Bi-CuDNNLSTM layer implemented in this study has a myriad of benefits in reducing the time and experimental costs for conducting new fast and slow cyclic tests in the future and results in a fast and accurate insight into hysteretic behavior of bridge piers.

• Journal of Korea Water Resources Association
• /
• v.53 no.2
• /
• pp.97-105
• /
• 2020

Recently, Artificial Neural Network receives attention as a data prediction method. Among these, a Long Shot-term Memory (LSTM) model specialized for time-series data prediction was utilized as a prediction method of hydrological time series data. In this study, the LSTM model was constructed utilizing deep running open source library TensorFlow which provided by Google, to predict inflows of multipurpose dams. We predicted the inflow of the Yongdam Multipurpose Dam which is located in the upper stream of the Geumgang. The hourly flow data of Yongdam Dam from 2006 to 2018 provided by WAMIS was used as the analysis data. Predictive analysis was performed under various of variable condition in order to compare and analyze the prediction accuracy according to four learning parameters of the LSTM model. Root mean square error (RMSE), Mean absolute error (MAE) and Volume error (VE) were calculated and evaluated its accuracy through comparing the predicted and observed inflows. We found that all the models had lower accuracy at high inflow rate and hourly precipitation data (2006~2018) of Yongdam Dam utilized as additional input variables to solve this problem. When the data of rainfall and inflow were utilized together, it was found that the accuracy of the prediction for the high flow rate is improved.