Self-configurable Mirror Servers for Automatic Adaptation to Service Demand Fluctuation

Various network services are becoming more and more important serving roles in the social infrastructure in the Internet. The more clients a service has, the more load the server has, which obviously lowers the quality of service. To balance load and maintain quality, multiple servers (mirror servers) are commonly provided that are able to provide access to the same service. However, it is difficult to estimate the required number of mirror servers in advance. To overcome this, we propose a new concept called self-configurable server groups and discuss its basic mechanism where all server nodes communicate with one another in a peer-to-peer style to adjust to the expected number of servers. We also demonstrate the effectiveness of the proposed mechanism via some simulations.

[1]  Israel Ben-Shaul,et al.  Dynamic Self Adaptation in Distributed Systems , 2000, IWSAS.

[2]  Israel Ben-Shaul,et al.  System support for dynamic layout of distributed applications , 1999, Proceedings. 19th IEEE International Conference on Distributed Computing Systems (Cat. No.99CB37003).

[3]  Ricardo Bianchini,et al.  Dynamic cluster reconfiguration for power and performance , 2003 .

[4]  Israel Ben-Shaul,et al.  Dynamic Adaptation and Deployment of Distributed Components In Hadas , 2001, IEEE Trans. Software Eng..

[5]  Israel Ben-Shaul,et al.  Dynamic layout of distributed applications in FarGo , 1999, Proceedings of the 1999 International Conference on Software Engineering (IEEE Cat. No.99CB37002).

[6]  Amin Vahdat,et al.  Managing energy and server resources in hosting centers , 2001, SOSP.

[7]  Luís Bernardo,et al.  Scalable Service Deployment Using Mobile Agents , 1998, Mobile Agents.