QUEUE LENGTH-BUSY TIME DISTRIBUTION OF WEB USERS DATA WITH SELF SIMILAR BEHAVIOR

It has been reported that Internet traffic exhibits self-similarity. Motivated by this fact, real time web users at different web centers can be treated as arrivals consider as traffic and has been examined by various techniques to test the self-similarity. The outcome from the experiments carried out and proven that arrival pattern has self similar behavior. In this paper, various techniques used to compute Hurst index which is a measure to know the intensity of self-similarity. In addition to this, the mean queue length and busy time distribution at a web center has been computed against traffic intensity. Numerical results clearly reveal that analysis accessible in this paper is very helpful in improvement of designs of web centers to give quality of service (QOS) guarantee.