Abstract How to competently apportion system resource to process the client request by gateway servers is a tricky predicament. In this paper, we propose an enhanced proposal for autonomous recital of gateway servers under highly dynamic traffic loads. We have developed G/G/1 queuing model algorithm and premeditated its intricacy, so that there is lossless information repossession at each node of gateway server. This facilitates to reduce response time variance in existence of bursty traffic. The most widespread contemplation is performance, because gateway servers must offer cost- effective and high-availability services in the elongated period, thus they have to be scaled to meet the expected load. performance measurements can be the base for performance modeling and prediction. With the help of performance models, the performance metrics (like buffer estimation, waiting time) can be determined at the development process, so that there is lossless information retrieval of data at every node of internet gateway servers. The paper portrays the assessment of buffer size using G/G/1 queuing model to estimate the final value of the memory size and then examine its implementation at the gateway servers. The obtained output is based on the simulation and experimental studies using synthesized workloads and analysis of real-world gateway servers demonstrate the effectiveness of the proposed system.

References

[1] Barford P. and Crovella M., Generating Representative Web Workloads for Network and Server Performance Evaluation, Measurement and Modeling of Computer Systems , pp. 151-160, 1998.

[2] Chen X. and Mohapatra P., Performance Evaluation of Service Differentiating Internet Servers, IEEE Transactions on Computers , vol. 51, no. 11, pp. 1368-1375, 2002.

[3] Kleinrock L., Queueing Systems, Applications, John Wiley, vol. 2, 1976.

[4] Lazowska E., Zahorjan J., Graham G., and Sevcik K., (Eds.) , Quantitative System Performance: Computer System Analysis Using Queueing Network Models , Prentice-Hall, Inc., 1984.

[5] Low S. and Srikant R., A Mathematical Framework for Designing a Low-Loss, Low- Delay Internet, IEEE Transactions on Communications , 2002.

[6] Menasce D. and Almeida V., Capacity Planning for Web Services: Metrics, Models, and Methods , Prentice Hall, 2001.

[7] Mogul J., Operating Systems Support for Busy Internet Servers, in the 5 th Workshop on Hot Topics in Operating Systems (HotOS-V) , Orcas Island, WA, 1995.

[8] Singh L. and Srivastava R., Estimation of Buffer Size of Internet Gateway Server via G/M/1 Queuing Model, International Journal of Applied Science, Engineering and Technology , vol. 4, no. 1, pp. 474-482, 2007.

[9] UDDI, Universal Description, Discovery, and Integration of Business for the Web , 2004.

[10] Ying L., Dullerud G., and Srikant R., Global Stability of Internet Congestion Controllers with Heterogeneous Delays, IEEE Transactions on Networking , vol. 14, no. 3, pp. 579-590, 2006.

[11] Zari M., Saiedain H., and Naeem M., Understanding and Reducing Web delays, IEEE Journal for Electronics and Computer Science , vol. 34, no. 12, pp. 30-37, 2001. Laxmi Singh is presently professor and head of Department of Solid State Physics and Electronics, Dr. RML. Avadh University, Faizabad, India. He is also the director of Institute of Engineering and Technology, University Campus, Dr. RML Avadh University, Faizabad. Riktesh Srivastava received his MS degree in electronics in 1998. Since 2004, he is working towards his PhD at the Department of Solid State Physics and Electronics of the Dr. RML Avadh University. His research is focused on internet server buffer estimation using queuing theory.