Scheduling Algorithms in Web Servers Clusters
نویسنده
چکیده
Abstract—Scheduling of jobs in web cluster of servers is a major research activity in Distributed Computing System (DCS). A key issue in server load balancing in a DCS is to select an effective load balancing scheme to distribute clients’ requests to the servers. This paper considers weighted round robin, shortest queue and diffusive load balancing policies. Performance of each of these policies was analyzed and compared. In this paper, we have investigated the problem of server load balancing and evaluated various server load balancing policies. The objective is to identify the techniques that produce good overall system performance.
منابع مشابه
A Content Aware Scheduling System for Network Services in Linux Clusters
With explosive growth of Internet, more and more companies are in need of powerful web servers to support e-commerce and other business activities. To meet this need, cluster architecture has emerged to be the most popular choice for high performance web servers. As one of the most important key technologies, content aware scheduling is becoming a hot research topic. Content-aware scheduling sy...
متن کاملA Classification of Job Scheduling Algorithms for Balancing Load on Web
Through this report, a classification of different job scheduling algorithms available for balancing the load on web servers is made. Types such as static and dynamic scheduling algorithms are thoroughly discussed and the strengths and weaknesses of these algorithms are put forth through this article.
متن کاملDeferred Assignment Scheduling in Clustered Web Servers
This paper proposes new scheduling policies for clustered servers, which are based on job size. The proposed algorithms are shown to be efficient, simple and easy to implement. They differ from traditional methods in the way jobs are assigned to back-end servers. The main idea is to defer scheduling as much as possible in order to make better use of the accumulated information on job sizes. Fur...
متن کاملDynamic Load Balancing in Geographically Distributed Heterogeneous Web Servers
With ever increasing Web traac, a distributed multi-server Web site can provide scalability and ex-ibility to cope with growing client demands. Load balancing algorithms to spread the requests across multiple Web servers are crucial to achieve the scalability. Various domain name server (DNS) based schedulers have been proposed in the literature, mainly for multiple homogeneous servers. The pre...
متن کاملFast Web Page Allocation On a Server Using Self- Organizing Properties of Neural Networks
This paper presents a novel competitive neural network learning approach to schedule requests to a cluster of Web servers. Traditionally, the scheduling algorithms for distributed systems are not applicable to control Web server clusters because different client domains have different Web traffic characteristics, Web workload is highly variable, and Web requests show a high degree of self-simil...
متن کامل