• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06a
• /
• pp.412-414
• /
• 2006

그리드 네트워크에서 MPI를 사용하여 지리적으로 산재된 컴퓨팅 자원을 활용하고 복잡한 문제를 해결하기 위한 MPICH-G2는 사설 IP 클러스터를 지원하지 못한다는 단점을 가지고 있다. MPICH-G2가 가지는 이러한 문제점을 해결하는 방법으로서 사설 IP 클러스터를 지원하는 MPICH-GP를 NAT와 프락시를 병용하여 설계하고 구현하였다. 사설 IP 클러스터의 프론트 노드에 프락시를 두고, 이를 통해 내부 계산 노드로의 통신 링크를 중계한다. 따라서 사설 IP 클러스터와 공인 IP 클러스터가 혼재된 네트워크에서도 적정한 경로를 설정하고 성공적으로 MPI 작업을 수행할 수 있다. 본 논문에서는 MPICH-GP의 성능을 기존의 MPICH-G2와 비교하였다. 그리드 환경에서 MPICH-GP는 MPICH-G2의 80% 이상의 성능을 보이며 rank 관리기법을 적용한 경우는 95% 이상의 성능을 나타낸다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.179-182
• /
• 2002

그리드 환경에서는 사용자가 컴퓨팅 자원을 사용하기 위해 시스템마다 로그인 할 필요가 없다. 어디서나 멀리 있는 자원으로 작업을 던지기만 하면 되는데, 이때 하나의 자원만을 사용하는 것이 아니라 여러 종류, 여러 개의 자원을 동시에 사용할 수 있다. 이러한 환경에서 각 사용자가 자원을 사용한 기록으로 남는 로그(log)처리는 기존의 계산 환경에서의 그것과는 많은 차이가 생기게 된다. 우선 사용자는 자신의 작업에 대한 로그를 보기 위해서 시스템마다 로그인 할 필요가 없어야 한다. 또 작업에 대한 효율적인 모니터링을 위해 정확한 로그를 필요로 하게 된다. 자원 관리자는 정확한 로그를 이용해 자원 관리 및 어카운팅(accounting) 정책을 수립할 수 있을 것이다. 본 논문에서는 이러한 그리드 환경에 적합한 로그 로그 생성 시스템을 설계하고자 한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.73-75
• /
• 2003

고속 네트워크의 등장으로 관리 영역을 초월한 계산 자원의 공유가 가능하게 되었고, 그리드 컴퓨팅이 등장하였다. 그리드 환경에 포함된 각 자원들은 이질적이고, 이질적인 환경에서 고성능을 얻기 위해서는 효과적인 자원 발견 및 자원 선택이 중요하다. 본 논문에서는 균등 분배 MPI 프로그램들에 대한 성능 예측 알고리즘과 자원 선택 알고리즘을 제안한다. 성능 예측 알고리즘은 자원의 이질성, 네트워크 성능, 복수 노드에 있는 CPU 부하, 응용 프로그램의 특성을 고려해 성능을 예측 한다. 자원 선택 알고리즘은 k개의 후보 자원 집합들을 생성한 후, 이들 중 최적의 집합을 선택한다. 이를 통해 기존 그리디 알고리즘의 약점이던 지역성을 극복했다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.12
• /
• pp.2695-2701
• /
• 2012

We need privacy protection protocols, that satisfy three essential security requirements; confidentiality, indistinguishability and forward security, in order to protect user's privacy in RFID system. The hash-chain based protocol that Ohkubo et. al proposed is the most secure protocol, that satisfies all of the essential security requirements, among existing protocols. But, this protocol has a disadvantage that it takes very long time to identify a tag in the back-end server. In this paper, we propose a scheme to keep security just as it is and to reduce computation time for identifying a tag in back-end server. The proposed scheme shows the results that the identification time in back-end server is reduced considerably compared to the hash-chain based protocol.

• The Journal of the Korea Contents Association
• /
• v.8 no.1
• /
• pp.137-142
• /
• 2008

In this paper, we describe TIGRIS Grid MPI Service, which is 4he WS-Resource Framework (WSRF) based services to enable an MPI job to be executed on Grid environments. It covers heterogeneous compute resources and diverse MPI libraries. The main functionalities are as follows. First, it allows an MPI user to seamlessly launch a job without knowing how to use the specific MPI library. Secondly, it executes an MPI job on the cross-site resources by supporting the Grid-enabled MPI library such as MPICH-G2. Thirdly, it enables the user to launch a job using the source code without compiling. The service is implemented on top of the services of Globus Toolkit. We provide the user interface as a web portal and CLI(Command Line Interface).

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.7
• /
• pp.380-389
• /
• 2006

In this paper, we present the MGrid system and its application for the construction of the Glycoconjugates simulation database called e-Glycoconjugates. The MGrid system is an integrated molecular simulation grid system for computing, databases, and analyses. For e-Glycoconjugates, we have been constructing the simulation database for 2,000 glycan chains and 100 glycoproteins until 2008. In this paper, we present the goal, architecture, and current implementation status of the MGrid system, and e-Glycoconjugates.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.12
• /
• pp.881-890
• /
• 2009

To solve big problem in high energy physics or bioinformatics, it needs a large number of computing resources. But it hard to be provided by one grid environment. While user can submit each job by using it's own user interface in each grid environment, it may need many cost and efforts to manage several hundred jobs conserved in each grid environment separately. In this paper, to solve this problem we develop Ganga's Gridway backend and InterGrid backend. Therefore as we provide the same grid user interface between different grid environments. We developed a Gridway backend module that provide users with access to globus-based grid resources as well. We have also developed an InterGrid backend that allows users to submit jobs that have access to both glite-based resources and globus-based resources. In order to demonstrate the practicality of the new backend plugin modules, we have integrated the AutoDock application used by WISDOM project into Ganga as a new built-in application plugin for Ganga. And we preformed interoperability experiment between PRAGMA grid based on Globus and EGEE grid based on gLite.

• Proceedings of the Korea Contents Association Conference
• /
• 2007.11a
• /
• pp.574-577
• /
• 2007

e-Science Service Integration Workbench is the core tool that enables IT-based computation and engineering researchers to collaborate their research activities via data sharing, by semi-automatically supporting the activities. Workbench provides the researchers with a scientific workflow by establishing the environment that is capable of finding, registering, composing, and executing services-legacy codes wrapped with grid services or web services-they need to use. In other words, designing their scientific workflow through the workbench, they can receive or share its final result with their collegues by submitting jobs they describe to computational resources such as supercomputers or grids or requesting an experiment. In this paper, we propose an implementation architecture of e-Science Service Integration Workbench to support grid services of Grid Middleware.

• The KIPS Transactions:PartA
• /
• v.14A no.1 s.105
• /
• pp.1-14
• /
• 2007

MPICH-G2 is an MPI implementation to solve complex computational problems by utilizing geographically dispersed computing resources in grid environments. However, the computation nodes in MPICH-G2 are exposed to the external network due to the lack of supporting the private IP clusters, which raises the possibility of malicious security attacks. In order to address this problem, we propose MPICH-GP with a new relay scheme combining NAT(Network Address Translation) service and an user-level proxy. The proxy running on the front-end system of private IP clusters forwards the incoming connection requests to the systems inside the clusters. The outgoing connection requests out of the cluster are forwarded through the NAT service on the front-end system. Through the connection path between the pair of processes, the requested MPI jobs can be successfully executed in grid environments with various clusters including private IP clusters. By simulations, we show that the performance of MPICH-GP reaches over 80% of the performance of MPICH-G2, and over 95% in ease of using RANK management method.

• Journal of the Korea Society for Simulation
• /
• v.19 no.2
• /
• pp.81-89
• /
• 2010

Biological data mining has been noticed as an issue as the volume of biological data is increasing extremely. Grid technology can share and utilize computing data and resources. In this paper, we propose a hybrid system that combines biological data mining with grid technology. Especially, we propose a decision range adjustment algorithm for processing efficiency of biological data mining. We obtain a reliable data mining recognition rate automatically and rapidly through this algorithm. And communication loads and resource allocation are key issues in grid environment because the resources are geographically distributed and interacted with themselves. Therefore, we propose a dynamic load balancing algorithm and apply it to the grid-based biological data mining method. For performance evaluation, we measure average processing time, average communication time, and average resource utilization. Experimental results show that this method provides many advantages in aspects of processing time and cost.