A Grid-Based Research Environment for Civil Engineering JANG HO LEE DONG WOOK KIM, SUN JANG Hongik University Dept. of Computer Engineering 72-1 Sangsu, Mapo, Seoul 121-791 KOREA janghol@cs.hongik.ac.kr {realtroy.kim, realjangsun}@gmail.com TAIKYEONG JEONG Myongji University Dept. of Communication Eng. San 38-2 Namdong, Cheoin-gu Yongin, Gyeonggido 449-728 KOREA ttjeong@mju.ac.kr YI SONG YI Seoul National University School of Computer Science and Eng. San 56-1, Sillim, Gwanak Seoul 151-742 KOREA yis@snu.ac.kr Abstract: The flow of information brings a tremendous change in the civil engineering industry as well as the politics, economy and culture of a society. The civil engineering field with the information revolution sees the combination of digital technology and construction technology, such as development of integrated construction systems, electronic databases, document management systems, and simulations. As a result, the scale of con- structed structures gets larger, multi-functional and complex. These core technologies and the development of extreme technology become an essential part of a nation’s competitive construction strategy. However, related huge experiment facilities are too expensive for an organization, which makes the building and application of them difficult. In this paper, we propose a grid-based research collaboratory that connects fragmented construction ex- periment facilities and shares their technologies and information across the research community, which ensures the efficiency of the facilities. And based on this research, we propose a design for a real-time hybrid experiment collaboratory in Korea Construction Engineering Development (KOCED) civil engineering research project and describe our experiment with its prototype. Key–Words: Grid system, Remote experiment, Collaborative computing, Civil engineering 1 Introduction In order to bring the construction technologies, such as experimentation, interpretation, design, building and maintenance up to a global level, we distribute the huge research facilities across the nation [8]. Also, al- though we need each research facility, we experience the limitation of accommodating the facilities to many research institutes. We share the facilities and maxi- mize the effectiveness of their use, through grid tech- nology [7, 6]. By distributing the research facilities across the nation, we can expect a balanced develop- ment of all the regions nation-wide. By installing the facilities into universities, we can expect the combi- nation of education and research. The KOCED grid system is huge research facilities, computing servers, and data repositories connected by a high speed infor- mation network and integrated to a grid system, which makes the facilities become one facility. This effort of distributing huge research facilities and building a grid system resemble nation-wide poli- cies of the U.S. and others. The Network for Earth- quake Engineering Simulation (NEES) project of the U.S. is one of the examples of such an effort [9, 3]. KOCED is composed of the cores of each region that are connected by Internet. In order for remote users to use the cores, it is necessary to have an entry point. This entry point provides a platform for facil- ities and services from the cores of each region. For this, this contains managing, monitoring and storage tools, and plays a role as an internet protocol gateway for user and the core of regions. Additionally, this pro- vides services and resource management for sharing data, and provides services for co-work. For an en- try point to access to experiment data, there should be an interface, and the data streaming technology makes this possible. Also, for remote control of physical ex- perimentation, it is required to have control protocol, and this protocol provides a standard interface for lo- cal facilities. The prototype of KOCED remote ex- perimental environment was built with Globus Toolkit 3 [5] which is based on OGSA [4]. Fig 1 shows the civil engineering experimental fa- cilities that are linked in the KOCED grid network across the nation. We plan to expand the grid cov- erage to include more experimental facilities. This paper is organized as follows. In section 2, we discuss related work where we discuss NEES and our KOCED. In section 3, we explain the major ser- vices of KOCEDgrid , which is the grid system used Proceedings of the 2007 WSEAS International Conference on Computer Engineering and Applications, Gold Coast, Australia, January 17-19, 2007 100