• The KIPS Transactions:PartD
• /
• v.9D no.3
• /
• pp.399-408
• /
• 2002

Component technology was proposed and applied to software development to overcome software crisis. Software component is a black box like an integrated circuit in hardware but it can not be utilized without good support specially for helping users understand efficiently. This paper shows that data book format for understanding hardware component can be well applied to representing software component. We selected an approach to understand component by matching the contents of data book with UML and API model technique. Besides, we added the architecture part and the interface which are the most important property of software component to the data book for software components. In order to verify effectiveness of components data book we extended batch descriptor in EJB and performed an experiment providing data book to programmers with components.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10b
• /
• pp.38-43
• /
• 2007

최근의 소프트웨어들이 다양한 기능을 갖추어가면서 점차 복잡도가 증가하고 있으며, 이에 따라 오류로부터의 복구도 어려워져 가고 있다. 이러한 변화는 소프트웨어의 자가 치유 연구에 중요한 이슈가 되고 있다. 하지만 자가 치유 방법론에서 중요한 요소 중에 하나인 모니터는 아직까지 개발자가 일일이 작성해야 하는 한계가 있다. 따라서 본 논문은 외부 자원으로 인한 오류를 탐지하는 모니터 모듈의 생성을 자동화하는 방법론을 제시하고, 이것을 적용한 소프트웨어 아키텍처를 제안한다. 본 방법론은 1) UML의 배치 다이어그램으로부터 소프트웨어와 하드웨어간의 연결을 분석하고, 2) 기술된 제약사항을 이용하여 모니터링 모듈을 자동으로 생성한다. 3) 이후 생성된 모듈을 소프트웨어 사양에 맞게 수정한 후 컴포넌트에 추가한다. 이러한 제안 방법론을 통해 기존에 수동으로 만들어야 했던 외부 자원 모니터를 자동화하는 것이 가능해 진다. 본 논문에서는 평가를 위해 제안 방법론을 비디오 회의 시스템의 클라이언트에 적용하여, 외부 자원의 오류를 올바르게 탐지해내는지 확인한다.

• Journal of Information Technology Services
• /
• v.7 no.4
• /
• pp.45-60
• /
• 2008

Networked enterprise (NE) is an organization of independent companies that collaborate with each other temporary or permanently for accomplishing common goals. The USA and EU have been developing principal concepts, techniques, and solutions to enhance the competitiveness of traditional industries including small-and-medium enterprises (SMEs). In Korea, however, implementation as well as R&D of NE is very few, which we believe comes from lack of understanding on Its meaning and lack of effective information systems for it. This paper is to suggest an Enterprise Architecture (EA) framework or reference model of NE and an Information Technology Architecture (ITA) of NE. The EA framework will help stakeholder of NE (e,g., policy makers, members of NE, IT solution providers, and researchers) understand structural and behavioral characteristics of NE. The ITA will be used as a guideline of developing information systems for NE that is essential for spreading networked business models, The focus of this paper is not on logical-level design but on conceptual-level modeling of NE. As verification of the suggested framework and architecture is still required, so we'll apply them to various manifestations of NE, e.g., dynamic supply chain, vertical integration of extended enterprises, and P2P-style virtual enterprises.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.164-165
• /
• 2015

고성능 복수 연산 처리 장치를 갖는 VLIW (Very Long Instruction Word)와 같은 프로세서 아키텍처는 정밀한 명령어 스케쥴링을 하드웨어가 아닌 소프트웨어가 처리해 주어야 한다. 통상 컴파일러가 하드웨어의 풍부한 자원을 충분히 활용할 수 있도록 이러한 기능을 수행하여 준다. 기존에 다양한 명령어 스케쥴링 알고리즘이 연구되었고 수 십년에 걸쳐 새로운 스케쥴링 기법들이 소개되었다. 이러한 스케쥴링 기법의 성능은 알고리즘의 효율뿐만 아니라 프로그램 코드에 내재된 의존관계 (dependence relation)의 복잡도에 따라 상당한 영향을 받는다. 본 연구에서는 의존도 완화기법으로서 레지스터 재할당 (register reallocation) 기법을 살펴보고 이를 활용하여 스케쥴링 성능 개선을 시도하여 보았다.

• Journal of Intelligence and Information Systems
• /
• v.14 no.2
• /
• pp.103-119
• /
• 2008

Traditional market which is characterized by the folksy retailing market has lost its competitiveness rapidly due to the emergence of the Internet and the change of customer's purchasing behavior. The recession of the traditional market contracts the regional economy. We suggest a u-Market, a traditional market with ubiquitous computing capability, to revitalize traditional market. The suggested u-Market system applies ubiquitous computing technologies characterized by communications between customers and objects without limitations of time and location. The proposed u-Market system offers location information and specific contents of traditional market to customers. Furthermore, u-Market system recommends the store and product list that customers are likely to visit and purchase based on their contexts, so they can save their time and effort to search the products or contents.

• Journal of Korean Society of Transportation
• /
• v.29 no.5
• /
• pp.33-42
• /
• 2011

As the scale of transfer center becomes gradually large, transfer information service considering convenience of users in motion in a large transfer center is required. This service provides moving users with information through individual portable terminals such as mobile phone or, smart phone which are capable of wireless communication. In particular, technology of measuring indoor location using Wi-Fi is in a stage of commercial practice and indoor/outdoor location-base transfer information service is possible through smart phones with built-in GPS and Wi-Fi. In this study, we suggest information services that support transfer activities based on the locations of visitors inside transfer centers. This study also suggests the logical and physical architectures, and physical components required for the flow of information as well as the target connected standardized information for connecting destinations of each proposed draft standards.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.2
• /
• pp.353-371
• /
• 2014

We already have seen many studies and articles about the methodology responding the security risks and threats. But we still have some controversial subjects to be settled. Now, we are living in the era that we should focus on how to use the security systems instead of how to make it. In this point of view, a company need to find out the answer for these questions, which security risks have to be handled in a corporate, which system is better for responding the security threats, and how we can build necessary security architecture in case of developing systems. In this article, we'd like to study on-site scenarios threatening the corporate assets, the limit on dealing with these threats, and how to consolidate the security events and information from enormous assets. Also, we'd like to search for the direction form the actual cases which have shown the desired effect from converging the assets and network informations.

• Journal of Korea Multimedia Society
• /
• v.11 no.2
• /
• pp.269-280
• /
• 2008

The information flow through RFID techniques such as, standardized EPC and EPCglobal network suggests new change in the method of cooperating and sharing information in the supply chain. Common function should be prepared for communication, security techniques, and business processes based on the EPCglobal standard to introduce efficient RFID techniques. And the standard architecture which can deal with the RFID environment flexibly is quite necessary. Therefore, this paper suggests Enterprise Application Framework (EAF) for developing efficient and secure RFID applications. EAF offers essential services to develop RFID enterprise systems : a communication environment which uses various standard communication protocols, a security function based on PKI, and abstract business services for flexible introduction of RFID businesses. It simplifies the RFID business integration, minimizes the development complexity, and thus EAF is expected to reduce the cost of RFID application development. EAF guarantees constructing secure RFID applications due to PKI-based security mechanism.

• Journal of Internet Computing and Services
• /
• v.6 no.4
• /
• pp.21-35
• /
• 2005

Software product-lines is the software development paradigm to attain the rapid development of quality applications by customizing the reconfigurable components and composing them based on predefined software architectures. Recently various methodologies for the component-based product lines are proposed, but these don't provide the specific implementation techniques of the components in terms of variability resolution mechanism. In other hand, the several approaches to implement the component supporting the variabilities resolution are developed, but these don't define the systematic analysis and design method considering the variabilities from the initial phase. This paper proposes the integration of PLUS, the one of product line methodologies extending UML modeling, and component code generation technique in order to increase the efficiency of producing the specific product in the software product lines. In this paper, the component has the hierarchical architecture consisting of the implementation elements, and each implementation elements are implemented as XSLT scripts. The codes of the components are generated from the feature selection. Using the microwave oven product lines as case study, the development process for the reconfigurable components supporting the automatic variability resolution is described.

• The KIPS Transactions:PartD
• /
• v.15D no.4
• /
• pp.569-580
• /
• 2008

Web services are the current most promising technology for service oriented architecture(SOA) implementations. However, in spite of the large scale acceptance of web services, they have been relegated to internal integration projects, and the grand vision of virtual enterprises where partners can be integrated on demand is yet to be realized. The main reason is that the current standards of web services are not very suitable for the dynamic web service discovery and integration. In this paper, we present a novel SemanticBPEL solution that merges the benefit of WS-BPEL, with the advantage of OWL-S for building a semi-automatic web service composition system. In particular, this work proposes a multi-phase search method for solving dynamic discovery and integration problems of web services. The proposed method is compared with the existing keyword based retrieval method. These comparisons show that our approach outperforms the existing method.