• Information and Communications Magazine
• /
• v.30 no.3
• /
• pp.43-50
• /
• 2013

현존하는 프로토콜 중심의 네트워킹이 가지는 근본적인 한계를 인식하고 이를 소프트웨어-정의 네트워킹(Software Defined Networking: SDN)이라는 새로운 흐름으로 해소하자는 노력이 확산되고 있다. 즉 실험자들이 점차 복잡해지는 네트워킹 문제들을 논리적으로 집중화된 단순함으로 재편하여 손쉽게 해결하자는 것이다. 본 논문에서는 SDN에 기반한 미래형 서비스 실증을 위해 핵심적인 도구로 대두된 시험환경(또는 테스트베드)의 구축 방향과 사례를 살펴본다. 먼저 가상화되고 프로그램이 가능한 융합형 실험자원들을 다수의 실험자들이 공용하는 환경을 구축하고, 개별적으로 서비스 실증을 자유롭게 시도하는 전체 프레임워크를 제시한다. 특히 융합형 자원을 Rack 방식으로 구성하고 이를 연동하여 시험환경을 구축하는 추세에 따라, 독자적인 SmartX Rack을 사용한 SDN 기반 서비스 실증을 위한 시험환경 구축 사례를 설명한다.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.47 no.7
• /
• pp.109-117
• /
• 2010

Building a large-scale testbed for Future Internet is very important to evaluate a new protocol and new network architecture designed by clean-slate approach. In Korea, new Future Internet testbed project, called FIRST (Future Internet Research for Sustainable Testbed), has been started since Mar. 2009 to design and test new protocols. This project is working together with ETRI and 5 universities. The FIRST@PC is to implement a virtualized hardware-accelerated PC-node by extending the functions of NetFPGA card and build a Future Internet testbed on the KOREN and KREONET for evaluating newly designed protocols and interesting applications. In this paper, we first briefly introduce FIRST@PC project and explain a 'MAC in IP Capsulator' user-space program using raw-socket in Linux to interconnect OpenFlow enabled switch sites on the KOREN and KREONET. After that, we address test results for TCP throughput performance for varying packet size. The test results show that the software based capsulator can support a reasonable bandwidth performance for most of applications.

• The Journal of the Convergence on Culture Technology
• /
• v.5 no.1
• /
• pp.401-407
• /
• 2019

SDN(Softeware Defined Networking) has emerged to address the rapidly growing demand for cloud computing and to support network virtualization services. Therefor many companies and organizations have taken SDN as a next-generation network technology. However, unlike the wired network where the SDN is originally designed, the SDN in the wireless network has a restriction that it can not provide the mobility of the node. In this paper, we extended existing openflow protocol of SDN and developed SDN-based network platform, which enables the SDN controller to manage the radio resources of its network and support the mobility of the nodes. The mobility support function of this paper has the advantage that a node in the network can move using its two or more wireless interfaces by using the radio resource management function of the SDN controller. In order to test the functions implemented in this paper, we measured parameters related to various transmission performance according to various mobile experiments, and compared parameters related to performance using one wireless interface and two interfaces. The SDN-based network platform proposed in this paper is expected to be able to monitor the resources of wireless networks and support the mobility of nodes in the SDN environment.

• KIISE Transactions on Computing Practices
• /
• v.23 no.7
• /
• pp.440-445
• /
• 2017

Due to the advance of e-commerce systems, the number of people using online shopping and products has significantly increased. Therefore, the need for an accurate recommendation system is becoming increasingly more important. Recurrent neural network is a deep-learning algorithm that utilizes sequential information in training. In this paper, an evaluation is performed on the application of recurrent neural networks to recommendation systems. We evaluated three recurrent algorithms (RNN, LSTM and GRU) and three optimal algorithms(Adagrad, RMSProp and Adam) which are commonly used. In the experiments, we used the TensorFlow open source library produced by Google and e-commerce session data from RecSys Challenge 2015. The results using the optimal hyperparameters found in this study are compared with those of RecSys Challenge 2015 participants.

• Journal of Digital Contents Society
• /
• v.19 no.8
• /
• pp.1545-1554
• /
• 2018

In order to support the mobility of the wireless devices, at least two APs (Access Points) must be arranged in a single AP area to maintain communication area. In the WLAN (Wireless LAN) environment, seamless handover is one of the most important issues in terms of effective utilization of wireless networks and maximization of services for users. On the other hand, SDN (Software-Defined Networking), which is emerging rapidly in recent years, is revolutionizing network management in terms of flexibility, fine control, and convenience. SDN originally reduces latency time or increases network robustness by real-time flow table control reducing or bypassing paths between switches in LAN-based data centers. In this study, we apply OpenFlow, a SDN platform focused on wired LAN, to a dense WLAN environment using legacy APs to implement and evaluate seamless handover for streaming services of digital contents.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.445-445
• /
• 2018

기후변화에 따른 환경오염은 21세기 인류에게 가장 심각한 문제 중의 하나로 대두되고 있다. 환경적인 측면에서 하천오염은 경제적으로 많은 문제를 발생시키고 있다. 이러한 하천오염 문제를 해결하기 위해서는 오염물질의 농도 측적 및 데이터 축적이 필수적이라 할 수 있다. 그러나 일반적으로 오염물질 부하량에 대한 직접적인 측정은 비용 측면에서 쉽지 않은 것이 사실이다. 또한 실시간으로 BOD, COD, TN, TP 등의 자료를 이용하여 예측하는 것에는 자료의 부족성으로 인해 한계가 있다. 본 연구에서는 구글의 딥러닝 오픈소스 라이브러리인 텐서플로우를 활용하여 기계학습을 통한 하천오염 예측을 목적으로 하고 있다. 기계학습을 위하여 텐서플로우를 활용하여 RNN, LSTM 인공신경망 모형을 구축하였다. 하천오염의 학습과 예측을 위해 결과치 분석을 위한 자료로는 금강 유역에 위치한 옥천 관측소 충청북도 옥천군 이원면 이원대교에 위치한 $36^{\circ}14'31.0''N$ $127^{\circ}40'02.6''E$의 관측소에서 BOD, COD, DO, 부유물질의 자료를 사용하였다. 모형의 학습을 위해서 입력자료는 수위, 유량, 평균기온, 평균풍속 자료를 2004년 ~ 2017년까지의 14년간의 자료를 사용하였다. 연구를 위해 BOD, COD, DO 부유물질 자료는 물환경정보시스템(http://water.nier.go.kr/)의 자료를 활용하고 수위, 유량등의 자료는 국가수자원관리종합정보시스템 (http://www.wamis.go.kr/)의 자료를 사용하였다. 그러나 수온, 수위, 풍속등의 자료는 일 자료가 있는가 반면 BOD, COD, TN, TP등의 자료는 일 자료가 있지 않아 이를 원활히 활용할 수 있도록 예측을 위한 결과치의 선형보간법을 통해 일 자료를 획득한 후 연구를 하였다. RNN, LSTM의 분석 시 학습속도, 반복시행횟수 sequence length의 길이 등의 값을 조절 하면서 결과치를 분석하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.10
• /
• pp.117-124
• /
• 2021

In this paper, we propose CANVAS (Creative ANalytics enVironment And System), an analytics system of the National Research Data Platform (DataON). CANVAS is a personalized analytics cloud service for researchers who need computing resources and tools for research data analysis. CANVAS is designed in consideration of scalability based on micro-services architecture and was built on top of open-source software such as eGovernment Standard framework (Spring framework), Kubernetes, and JupyterLab. The built system provides personalized analytics environments to multiple users, enabling high-speed and large-capacity analysis by utilizing high-performance cloud infrastructure (CPU/GPU). More specifically, modeling and processing data is possible in JupyterLab or GUI workflow environment. Since CANVAS shares data with DataON, the research data registered by users or downloaded data can be directly processed in the CANVAS. As a result, CANVAS enhances the convenience of data analysis for users in DataON and contributes to the sharing and utilization of research data.

• Journal of Korean Society of Archives and Records Management
• /
• v.18 no.2
• /
• pp.163-188
• /
• 2018

How do we create a next generation Records Management System? Under a rapidly changing system development environment, the records management system of public institutions has remained stable for the past 10 years. For this reason, it seems to be the key cause of the structural problem of the Records Management System, which makes it difficult to accommodate user requirements and apply a new recording technology. The purpose of this study is to present a redesigned model for a next-generation records management system by analyzing the status of the electronic records management. This study analyzed "A Study on the Redesign of the Next-Generation Electronic Records Management Process," records management technology of advanced records management system, and a case of an overseas system. Based on the analysis results, the improvement direction of the records management system was analyzed from several aspects: functional, software design, and software distribution. This study thus suggests that the creation of a microservice architecture-based (MSA) and an open source software-oriented (OSS) records management system should be the focus of next-generation record management.

• KIISE Transactions on Computing Practices
• /
• v.21 no.11
• /
• pp.727-732
• /
• 2015

These days, many researchers are working on Future Internet and a new networking paradigm called Software Defined Networking draws a great attention. In this paper, we redefine Software Defined Networking as Service Defined Networking which means that packets are categorized according to types of services. By using Service Defined Networking, we are not only dealing with the way to utilize the network resources efficiently but we also propose an algorithm to minimize the waiting time for packets to be delivered. This proposed algorithm can solve the delay problem, one of the most significant problems caused by network congestion. Also, since we are adopting Service Defined Networking, network resource utilization can be improved compared to the existing network.

• Journal of the Korea Convergence Society
• /
• v.9 no.9
• /
• pp.431-440
• /
• 2018

In this study, authors predicted probability of resignation of newly employed nurses using TensorFlow, an open source software library for numerical computation and machine learning developed by Google, and suggested strategic human resources management plan. Data of 1,018 nurses who resigned between 2010 and 2017 in single university hospital were collected. After the order of data were randomly shuffled, 80% of total data were used for machine leaning and the remaining data were used for testing purpose. We utilized multiple neural network with one input layer, one output layer and 3 hidden layers. The machine-learning algorithm correctly predicted for 88.7% of resignation of nursing staff with in one year of employment and 79.8% of that within 3 years of employment. Most of resigned nurses were in their late 20s and 30s. Leading causes of resignation were marriage, childbirth, childcare and personal affairs. However, the most common cause of resignation of nursing staff with in one year of employment were maladaptation to the work and problems in interpersonal relationship.