Client-class based admission control algorithm in Web server using cluster analysis
Hea-Sook Park, Eun-Kyung Jung, Doo-Kwon Baik · 2005
This paper analyzes the level of contribution vis-a-vis the clients' service patterns in multimedia contents services and profits for consolidating a company's profits through the cluster analysis of clients' data applying the TwoStep Cluster Method of CRM. This paper proposes a new algorithm for allocating the media server's resources selectively by clusters using the cluster analysis. Clients were grouped into 4 clusters according to the contribution level in terms of profits. Likewise, the CRFS (client request filtering system) was suggested to allocate media server resources per cluster. CRFS issues admission within the resource limit of the cluster where the client belongs.