• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.5
• /
• pp.265-273
• /
• 2002

With the explosive growth of WWW traffic, there is an increasing demand for the high performance Web servers to provide a stable Web service to users. The cluster-based Web server is a solution to core with the heavy access from users, easily scaling the server according to the loads. In the cluster-based Web sewer, a back-end node may not be able to serve some HTTP requests directly because it does not have the requested contents in its main memory. In this case, the back-end node has to retrieve the requested contents from its local disk or other back-end nodes in the cluster. To reduce service latency, we introduce a new prefetch scheme. The back-end nodes predict the next HTTP requests and prefetch the contents of predicted requests before the next requests arrive. We develop three prefetch algorithms bated on some useful information gathered from many clients'HTTP requests. Through trace-driven simulation, the service latency of the prefetch scheme is reduced by 10 ~ 25% as compared with no prefetch scheme. Among the proposed prefetch algorithms, Time and Access Probability-based Prefetch (TAP2) algorithm, which uses the access probability and the inter-reference time of Web object, shows the best performance.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.12
• /
• pp.708-713
• /
• 2003

For many years clustered heterogeneous web server architecture has been formed on the internet because the explosive internet services and the various quality of requests. The critical point in cluster environment is the mapping schemes of request to server. and recently this is the main issue of internet architecture. The topic of previous mapping methods is to assign equal loads to servers in cluster using the number of requests. But recent growth of various services makes it hard to depend on simple load balancing to satisfy appropriate latency. So mapping based on requested content to decrease response time and to increase cache hit rates on entire servers - so called “content-based” mapping is highly valuated on the internet recently. This paper proposes Prioritized Highest Random Weight mapping(PHRW mapping) that improves content-based mapping to properly fit in the heterogeneous environment. This mapping scheme that assigns requests to the servers with priority, is very effective on heterogeneous web server cluster, especially effective on decreasing latency of reactive data service which has limit on latency. This paper have proved through algorithm and simulation that proposed PHRW mapping show higher-performance by decrease in latency.

• The Journal of the Korea Contents Association
• /
• v.6 no.9
• /
• pp.9-18
• /
• 2006

We completed a new cluster system based on WDM by proposing a virtual topology reconfiguration schemes. The key idea of the proposed scheme is to construct a set with the longest chains of requests of connecting nodes which need to be assigned a wavelength. All the sets have no common factor, that is, there is no duplicated link among the requests of connecting. After making the set satisfying this condition, we could assign a wavelength to per corresponding set. We could reconfigure a virtual topology with this scheme collectively. we compared our scheme to existing approaches by the OWns simulation tool. As the results, we gained improved performances, reducing 10% of blocking rate and improving 30% of ADM utilization in terms of the blocking probability and the ADM utilization.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.9
• /
• pp.1145-1157
• /
• 1999

본 논문에서는 RFC 2022에서 제안한 MARS 모델을 기반으로 하여 단일 대규모 클러스터를 가지는 ATM 망에서 다중의 멀티캐스팅 서버(MCS)를 이용한 멀티캐스팅 방안을 제안하고 그 성능을 평가하였다. 클러스터 내의 한 ATM 호스트가 특정 IP 멀티캐스트 그룹에 가입할 경우 ATM 호스트의 위치와 이미 존재하는 멀티캐스팅 서버들 사이의 전송 지연을 고려하여 가능한 한 종단간 전송 지연을 최소화하는 멀티캐스팅 서버를 선택하는 방안을 기술하였다. 이 방안은 최단거리 경로 알고리즘(shortest path algorithm)에 기반하여 최적의 MCS를 선정하고 송수신자 사이의 최소 지연을 가지는 멀티캐스트 트리를 구성한다. 다양한 망 위상에서 MCS의 분포 패턴을 다르게 할 경우에도 이 방안은 멀티캐스트 트리의 평균 전달 지연을 줄이는 것을 시뮬레이션을 통하여 확인하였다.Abstract In this paper, we proposed a scheme to support multiple MCSs over a single and large cluster in ATM networks, evaluated its performance by simulation. When an ATM host requests joining into a specific multicast group, the MARS designate a proper MCS among the multiple MCSs for the group member to minimize the average path delay between the sender and the group members. This scheme constructs a multicast tree through 2-phase partial multicast tree construction based upon the shortest path algorithm.We reduced the average path delay in multicast tree using our scheme under various cluster topologies and MCS distribution scenarios.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.12
• /
• pp.648-654
• /
• 2002

The Virtual Interface Architecture(VIA) is an industry standard for communication over system area networks(SANs). M-VIA is a software implementation of VIA technology on Linux. In this paper, we implemented the M-VIA on an AceNIC Gigabit Ethernet by developing a new AceNIC driver for the M-VIA. We analyzed the M-VIA data segmentation processes. When a Gigabit Ethernet MTU is larger than 1514 bytes, M-VIA data segmentation size leaves much room for improvement. So we experimented with various MTU and M-VIA data segmentation size and compared the performances.

• Journal of the Korean Academic Society of Industrial Cluster
• /
• v.1 no.1
• /
• pp.47-66
• /
• 2007

The progress of digital revolution leads to technology and industry infusion which make new products and new commodity markets in the existing industries. Under the circumstance of new innovation patterns in industries and globalization of the world, innovation firms adapt different evolution processes for different business situations. Technological innovation enhances the competitiveness of fins in the business world. This paper shows the researches on technological innovation which make constant economy growth in the knowledge based economy. And this paper also suggests the policy on technological innovation for the future growth of society. Main focus of this paper are as follows. First, the theoretical survey on technological innovation is reviewed. Second, the program and the supporting system for technological innovation in the USA and Europe are investigated. Third, the situation and problems of Korea is analysed. Fourth, strategy and suggestions on the development of technological innovation in the 21st century knowledge based economy is suggested through the window of cluster and regional economy development.

• Journal of KIISE:Computer Systems and Theory
• /
• v.27 no.5
• /
• pp.496-505
• /
• 2000

Clusters of workstations (COW) are becoming an attractive option for parallel scientific computing, a field formerly reserved to the MPPs, because their cost-performance ratio is usuallybetter than that of comparable MPPS, and their hardware and software can be easily enhanced to thelatest generations. In this paper we present the design and implementation of our runtime library forclusters of workstations, called "Collective I/O Clustering". The library provides a friendlyprogramming model for the I/O of irregular applications on clusters of workstations, being completelyintegrated with the underlying communication and I/O system. In the collective I/O clustering, two I/Oconfigurations are possible. In the first I/O configuration, all processors allocated can act as I/Oservers as well as compute nodes. In the second I/O configuration, only a subset of processors canact as I/O servers, The compression and software caching facilities have been incorporated into thecollective 1/0 clustering to optimize the communication and I/O costs. All the performance results wereobtained on the IBM-SP machine, located at Argonne National Labs.

• Journal of the Economic Geographical Society of Korea
• /
• v.8 no.2
• /
• pp.217-236
• /
• 2005

This paper aims to analysis the production linkage relationships between large firm and its subcontracting firms, especially focus on the spatial network and collaboration network. For the purpose of it, the division of semiconductor in Samsung Electronics and its most important subcontracting firms are picked up for the case study. The empirical study show that the location distribution pattern of Samsung Electronic's subcontracting firms is concentrated very highly on the Kyeong-Ki Province and Chung-Nam Province in Korea, which is the location sites of Samsung Electronic's key plants as well as the best environment for business in Korea. And the major subcontracting firms seems to be in the hierarchic and vertical relationship with Samsung Electronics rather than horizontal and good collaboration network in this case.

• Journal of KIISE:Computer Systems and Theory
• /
• v.29 no.10
• /
• pp.582-588
• /
• 2002

Scalable displays generate large and high resolution images and provide an immersive environment. Recently, scalable displays are built on the networked clusters of PCs, each of which has a fast graphics accelerator, memory, CPU, and storage. However, the distributed rendering on clusters is a network bound work because of limited network bandwidth. In this paper, we present a new algorithm for reducing the network bandwidth and implement it with a conventional distributed rendering system. This paper describes the algorithm called geometry tracking that avoids the redundant geometry transmission by indexing geometry data. The experimental results show that our algorithm reduces the network bandwidth up to 42%.

• Composites Research
• /
• v.13 no.1
• /
• pp.50-60
• /
• 2000

This paper presents an analytical method considering tensile strength enhancement in hybrid $Al_2O_3$ fiber/particle/aluminum composites(MMCs). The tensile strength and elastic modulus of the hybrid MMCs are even 20% higher than those of the fiber reinforced MMCs with same volume fraction of reinforcements. This phenomenon is explained by the cluster model which is newly proposed in this research, and the strengthening mechanisms by a cluster is analyzed using simple modified rule of mixtures. From the analysis, it is observed that cluster structure in hybrid MMCs increase the fiber efficiency factor for the tensile strength and the orientation factor for the elastic modulus. The present theory is then compared with experimental results which was performed using squeeze infiltrated hybrid MMCs made of hybrid $Al_2O_3$ short fiber/particle preform and AC8A alloy as base metal, and the agreement is found to be satisfactory.