• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.16-18
• /
• 2003

클러스터 파일 시스템은 데이터 입출력 대역폭을 극대화하여 효율성을 높이고 각 노드의 입출력 부담을 균등하게 부과하기 위하여 원본 파일을 여러 노드에 분산 저장한다. 이렇게 파일을 노드들에 분산 저장하기 위해서는 효율적인 노드간 데이터 통신을 필요로 하며, 노드 내부에서도 클러스터 파일 시스템과 어플리케이션과의 효율적인 전용 데이터 교환 메커니즘을 지원해야 한다. 본 논문에서는 WAN(Wide Area Network)에 적합하도록 설계된 TCP를 이용한 기존의 교환 메커니즘인 통신모듈이 가지고 있는 문제점을 해결하기 위해 다양한 프로토콜과 하드웨어적인 접근을 통해 Myrinet이 초고속 통신모듈에 적합함을 보이고 GM API를 활용하여 기존의 소켓기반인 TCP/IP를 이용한 통신 모듈을 대처할 새로운 통신모듈의 모델을 제시한다.

• Journal of KIISE:Information Networking
• /
• v.32 no.1
• /
• pp.68-79
• /
• 2005

The popularity of Web service is ever increasing. As the number of services and clients continue to increase, the problem of providing a system that scales with this increase is becoming more difficult. A costly and ineffective method is to buy a new system that is more powerful every time the load becomes unbearable. h more cost effective solution is to expand the system as the need arises. This is the approach taken in Web cluster systems. However, providing effective scalability in a Web cluster system is stil1 an open issue. In this study, we implement a Web cluster system based on Layer 7 switching technique on Linux. The implementation is based on a design proposed and implemented by Aron et al., but on the FreeBSD. Though the design li the same, due to the vast difference between the FreeBSD and Linux, the implementation presented in this paper is totally new. We also propose the Dual Scheduling (DS) load distribution algorithm that distributes the requests to the system resources by observing the variations in the request rate. We show through measurement on our implementation that the DS alorithm performs considerably bettor than previous algorithms.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10c
• /
• pp.718-720
• /
• 2002

본 논문에서는 대용량 공유 파일시스템의 자원을 효율적으로 관리할 수 있는 볼륨 관리기에 대해서 제안하고 리눅스 상에 구현을 해 보았다. SANfs[5]는 Storage Area Network(SAN)의 대용량 저장장치를 지원할 수 있도록 제안하고 구현된 확장성 있는 공유 파일 시스템이다. SANfs의 자원을 효율적으로 이용하기 위해서 저장장치들을 효율적으로 관리할 수 있는 도구가 필요하게 되었고, 이 논문에서 SANfs의 구조에 적합한 볼륨 관리기인 SANfs-VM을 새롭게 제안하고 구현하였다. SANfs-VM은 SANfs의 엔터프라이즈 컴퓨팅을 위해서 다양한 레벨의 RAID, online /resizing/reconfiguration 등의 기능을 제공함으로써 SANfs 저장장치의 확장성, 가용성을 향상시켰다. 또한 SANfs-VM은 저장 장치 시스템의 관리를 쉽게 함으로써 easy management 기능을 증진시켰다.

• The Journal of Society for e-Business Studies
• /
• v.9 no.1
• /
• pp.83-104
• /
• 2004

Recently, information systems that require storage and retrieval of huge amount of data are becoming used widely. Accordingly, research efforts have been made to develop Linux cluster file systems in the SAN environment in which clients themselves can manage metadata and access data directly. Also a semi-flat directory structure based on extendible hashing has been proposed to support fast retrieval of files[1]. In this research, we have designed and implemented the semi-flat extendible hash directory under the Linux system. In order to evaluate the practicality of the directory, we have also implemented the B+-tree based directory and experimented the performance. According to the performance comparisons, the extendible hash directory has the better performance at insert, delete, and search operations. On the other hand, the B+-tree directory is better at sorting files.

• Proceedings of the KSR Conference
• /
• 2008.06a
• /
• pp.1773-1780
• /
• 2008

When the fire occur in the deeply underground subway station, the difficulties of passenger evacuation are expected because of many stairs to the exit. In this study, SOONGSIL-University station (7 line, 47m depth) is the one of the deepest subway stations of the each line in the Seoul metro. The numerical computational-simulation was performed for the fire driven flow in the subway station. Hot and smoke flow was analyzed from the simulation results. The proper plan of evacuation against fire was considered through the results. The fire driven flow was simulated using FDS code in which LES method was applied. The Heat Release Rate was 10MW and the ultrafast model was applied for the growing model of the fire source. The proper mesh size was determined from the characteristic length of fire size. The parallel computational method was employed to compute the flow and heat eqn's in the meshes, which are about 10,000,000, with 6cpu of the linux clustering machine.

• Journal of the Korean earth science society
• /
• v.43 no.1
• /
• pp.60-76
• /
• 2022

This study presents a software full setup and the following test execution times in a Linux cluster for the United Kingdom Earth System Model (UKESM) and then compares the model results from control and experimental simulations of the UKESM relative to various observations. Despite its low resolution, the latest version of the UKESM can simulate tropospheric chemistry-aerosol processes and the stratospheric ozone chemistry using the United Kingdom Chemistry and Aerosol (UKCA) module. The UKESM with UKCA (UKESM-UKCA) can treat atmospheric chemistryaerosol-cloud-radiation interactions throughout the whole atmosphere. In addition to the control UKESM run with the default CMIP5 SO₂ emission dataset, an experimental run was conducted to evaluate the aerosol effects on meteorology by changing atmospheric SO₂ loading with the newest REAS data over East Asia. The simulation period of the two model runs was 28 years, from January 1, 1982 to December 31, 2009. Spatial distributions of monthly mean aerosol optical depth, 2-m temperature, and precipitation intensity from model simulations and observations over East Asia were compared. The spatial patterns of surface temperature and precipitation from the two model simulations were generally in reasonable agreement with the observations. The simulated ozone concentration and total column ozone also agreed reasonably with the ERA5 reanalyzed one. Comparisons of spatial patterns and linear trends led to the conclusion that the model simulation with the newest SO₂ emission dataset over East Asia showed better temporal changes in temperature and precipitation over the western Pacific and inland China. Our results are in line with previous finding that SO₂ emissions over East Asia are an important factor for the atmospheric environment and climate change. This study confirms that the UKESM can be installed and operated in a Linux cluster-computing environment. Thus, researchers in various fields would have better access to the UKESM, which can handle the carbon cycle and atmospheric environment on Earth with interactions between the atmosphere, ocean, sea ice, and land.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.151-154
• /
• 2002

갈수록 증가하는 인터넷 사용자의 서비스 요구량에 대처하기 위해, 부하 분산 기능을 갖는 클러스터 시스템의 이용이 늘어가고 있다. 본 연구에서는 클라이언트에게 보다 향상된 응답 성능을 제공하기 위해 사용되는 RR(Round Robin), WRR(Weighted Round Robin), LC(Least Connection), WLC(Weighted Least Connection) 부하 분산 알고리즘에 대해서, 클라이언트로부터 인입되는 7 가지 요구 수신 패턴에 따른 부하 분산 응답 특성을 분석하고 그 결과를 논한다. 이를 위해, 실제 시스템의 측정 결과를 토대로 단위 시간 당 인입되는 클라이언트의 요구량 변화를 7 가지 패턴으로 분류하였고, 리눅스 가상 서버(LVS: Linux Virtual Server) 클러스터 시스템을 대상으로 7 가지 요구 패턴에 대한 부하 분산 응답 특성을 얻었다. 본 연구를 통해서 클라이언트 요구랑 변화 패턴에 따른 최적의 부하 분산 알고리즘을 제시할 수 있었다. 본 연구 결과는 향후 효율적인 동적 부하 분산 연구에 좋은 참고가 될 것이다.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.6 s.44
• /
• pp.69-78
• /
• 2006

In the field of the bioinformatics, it plays an important role in predicting functional information or structure information to search similar sequence in biological DB. Biolrgical sequences have been increased dramatically since Human Genome Project. At this point, because the searching speed for the similar sequence is highly regarded as the important factor for predicting function or structure, the SMP(Sysmmetric Multi-Processors) computer or cluster is being used in order to improve the performance of searching time. As the method to improve the searching time of BLAST(Basic Local Alighment Search Tool) being used for the similarity sequence search, We suggest the nBLAST algorithm performing on the cluster environment in this paper. As the nBLAST uses the MPI(Message Passing Interface), the parallel library without modifying the existing BLAST source code, to distribute the query to each node and make it performed in parallel, it is possible to easily make BLAST parallel without complicated procedures such as the configuration. In addition, with the experiment performing the nBLAST in the 28 nodes of LINUX cluster, the enhanced performance according to the increase in the number of the nodes has been confirmed.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.9
• /
• pp.407-415
• /
• 2007

In parallel cluster computing systems, the efficiency of communication between computing nodes is one of important factors that decide overall system performance. Accordingly, many researchers have studied on high-performance inter-node communication. The recently launched multi-core processor, however. increases the importance of intra-node communication as well because the more the number of cores in a node, the more the number of parallel processes running in the same node. Though there have been studies on intra-node communications, these have limited considerations on the state-of-the-art systems. In this paper, we propose a Linux kernel module that minimizes the number of data copy by exploiting the memory mapping mechanism for high-performance intra-node communication. The proposed kernel module supports the Linux kernel version 2.6. The performance measurements over a multi-core system present that the proposed kernel module can achieve lower latency up to 62% and higher throughput up to 144% than an existing kernel module approach. In addition, the measurements reveal that the performance of intra-node communication can vary significantly based on whether the cores that run the communication processes are belong to the same processor package (i.e., sharing the L2 cache).

• Journal of KIISE:Information Networking
• /
• v.33 no.3
• /
• pp.231-246
• /
• 2006

A server load balancer is used to accept and distribute client requests to one of servers in a wireless internet proxy server cluster. LVS(Linux Virtual Server), a software based server load balancer, can support several load balancing algorithms where client requests are distributed to servers in a round robin way, in a hashing-based way or in a way to assign first to the server with the least number of its concurrent connections to LVS. An improved load balancing algorithm to consider server performance was proposed where they check upper and lower limits of concurrent connection numbers to be allowed within each maximum server performance in advance and apply the static limits to load balancing. However, they do not apply run-time server load information dynamically to load balancing. In this paper, we propose a dynamic load balancing scheme where the load balancer keeps each server CPU load information at run time and assigns a new client request first to the server with the lowest load. Using a cluster consisting of 16 PCs, we performed experiments with static content(image and HTML). Compared to the existing schemes, experimental results show performance improvement in the cases of client requests requiring CPU-intensive processing and a cluster consisting of servers with difference performance.