• ETRI Journal
• /
• v.31 no.2
• /
• pp.173-181
• /
• 2009

The growth of the World Wide Web and web-based applications is creating demand for high performance web servers to offer better throughput and shorter user-perceived latency. This demand leads to widely used cluster-based web servers in the Internet infrastructure. Load balancing algorithms play an important role in boosting the performance of cluster web servers. Previous load balancing algorithms suffer a significant performance drop under dynamic and database-driven workloads. We propose an estimation-based load balancing algorithm with admission control for cluster-based web servers. Because it is difficult to accurately determine the load of web servers, we propose an approximate policy. The algorithm classifies requests based on their service times and tracks the number of outstanding requests from each class in each web server node to dynamically estimate each web server load state. The available capacity of each web server is then computed and used for the load balancing and admission control decisions. The implementation results confirm that the proposed scheme improves both the mean response time and the throughput of clusters compared to rival load balancing algorithms and prevents clusters being overloaded even when request rates are beyond the cluster capacity.

• International Journal of Contents
• /
• v.4 no.1
• /
• pp.16-19
• /
• 2008

In this paper, we propose to new dispatcher method, which doesn't depend on an operating system of the server, and the direct routing method, by which a server answers a client's request at first hand. And, propose new web clustering scheme based on the contents on the web where web servers composed of cluster, with each different contents, answer client's request. The other purposes are to reduce overhead of the dispatcher through load balance, and to minimize the time to take in responding to a client's request. The performance of new web cluster scheme was improved by about 39% than that of the existing RR method. It was identified that the performance of the proposed web cluster method was extraordinary improved comparing with that of the existing RR method as a whole.

• The KIPS Transactions:PartC
• /
• v.9C no.1
• /
• pp.115-122
• /
• 2002

This paper presents a new web cluster scheme based on dispatcher which does not depend on operating system for server and can examine server's status interactively. Two principal functions are proposed for new web cluster technique. The one is self-controlled load distribution and the other is transaction fail-safe. Self-controlled load distribution function checks response time and status of servers periodically, then it decides where the traffic goes to guarantee rapid response for every query. Transaction fail-safe function can recover lost queries including broken transaction immediately from server errors. Proposed new web cluster scheme is implemented by C language on Unix operating system and compared with legacy web cluster products. On the comparison with broadcast based web cluster, proposed new web cluster results higher performance as more traffic comes. And on the comparison with a round-robin DNS based web cluster, it results similar performance at the case of traffic processing. But when the situation of one server crashed, proposed web cluster processed traffics more reliably without lost queries. So, new web cluster scheme Proposed on this dissertation can give alternative plan about highly increasing traffics and server load due to heavy traffics to build more reliable and utilized services.

• Communications for Statistical Applications and Methods
• /
• v.12 no.1
• /
• pp.191-197
• /
• 2005

Application of data mining techniques to the world wide web, referred to as web mining, has been the focus of several recent researches. With the explosive growth of information sources available on the world wide web, it has become increasingly necessary to track and analyze their usage patterns. In this study, we introduce a process of pre-processing and cluster analysis on web log data and suggest a distance measure considering the structural relationships between web contents. Also, we illustrate some real examples of cluster analysis for web log data and look into practical application of web usage mining for eCRM.

• The KIPS Transactions:PartA
• /
• v.8A no.3
• /
• pp.261-268
• /
• 2001

With the rapid growth of the Internet, various web-based businesses are creating a new environment in an imaginary space. However, this expanding Internet and user increase cause an overflow of transmission and numerous subordinate problems. To solve these problems, a parallel cluster system is produced using different methods. This thesis recommends a multi0distribution cluster group. It constructs a MPP dynamic distribution sub-cluster group using numerous low-priced and low-speed systems. This constructed sub-cluster group is then connected with a singular virtual IP to finally serve the needs of clients (users). This multi-distribution cluster group consists of an upper structure based on LVS and a dynamic serve cluster group centered around an SC-server. It conducts the workloads required from users in a parallel process. In addition to the web service, this multi-distribution cluster group can efficiently be utilized for the calculations which require database controls and a great number of parallel calculations as well as additional controls with result from the congestion of service.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.5 no.1
• /
• pp.23-32
• /
• 2009

Although the web is becoming a widely accepted medium, it provides relatively poor performance and low availability. A cluster consists of a collection of interconnected stand-alone computers working together and provides a high-availability solution in application area such as web services or information systems. Web server clusters require a high-availability service with a proactive and practical fault management. However, as the system complexity grows, it is not easy to meet the requirement. Therefore, web server clusters must have self-fault management capability for meeting high-availability requirement. In this paper, we propose high availability web server clusters using self-healing technique with a minimal human intervention. Our experimental results show that a proposed method can be used to improve the availability of web server clusters.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.8
• /
• pp.11-18
• /
• 2011

The increasing of web users load to the excessive transmission traffic and system overload problems. To solve these problems, cluster systems are studied. In conventional cluster systems, when the request size is large owing to such types as multimedia and CGI, the particular server load and response time tend to increase even if the overall loads are distributed evenly. In this paper, we consider the methodology of efficient resource usage, specially distributed web cluster system. We develope an algorithm that distributes the load on the web cluster system to use the system resources, such as system memory equally. The response time is chosen as a performance measure on the various clustering models. And based on the concurrent user to the web cluster system, the response time is also examined as the number of users increases. Simulation experience with this algorithm shows that the response time and average throughput seems to have a good results compare to those with the other algorithm.

• Journal of the Korea Society for Simulation
• /
• v.22 no.3
• /
• pp.71-79
• /
• 2013

Web server cluster systems are widely used, where a large number of PC level servers are interconnected via network. This paper focuses on forecasting an appropriate number of web servers which can serve four different classes of user requests, simple web page viewing, knowledge query, motion picture viewing and motion picture uploading. Two ways of serving different classes of web service requests are considered, commonly used web servers and service dedicated web servers. Computer simulation experiments are performed in order to find a good way of allocating web servers among different classes of web service requests, maintaining certain levels of resource utilization and response time.

• Communications for Statistical Applications and Methods
• /
• v.18 no.4
• /
• pp.549-557
• /
• 2011

In this paper, we apply cluster analysis to "Okayama parking data" that is one of the spatial point patterns data that includes locations and the fare structure of car parking space in Okayama central area. This study classifies the characteristics of small areas through Okayama parking data as well as visualizes the results of the cluster analysis. We develop web applications that connect the results of a cluster analysis and overlay objects including points of balloons and rectangles of small areas over a map of Okayama central area.

• Journal of Korea Spatial Information System Society
• /
• v.6 no.1 s.11
• /
• pp.3-17
• /
• 2004

With the rapid growth of the Internet geographic information services through the WWW such as a location-based service and so on. Web GISs (Geographic Information Systems) have also come to be a cluster-based architecture like most other information systems. That is, in order to guarntee high quality of geographic information service without regard to the rapid growth of the number of users, web GISs need cluster-based architecture that will be cost-effective and have high availability and scalability. This paper proposes the design of the cluster-based web GIS with high availability and scalability. For this, each node within a cluster-based web GIS consists of main memory spatial databases which accomplish role of caching by using data declustering and the locality of spatial query. Not only simple region queries but also the proposed system processed spatial join queries effectively. Compare to the existing method. Parallel R-tree spatial join for a shared-Nothing architecture, the result of simulation experiments represents that the proposed spatial join method achieves improvement of performance respectively 23% and 30% as data quantity and nodes of cluster become large.