Queue response to input correlation functions: discrete spectral analysis
IEEE/ACM Transactions on Networking (TON)
On variations of queue response for inputs with the same mean and autocorrelation function
IEEE/ACM Transactions on Networking (TON)
Wide-area Internet traffic patterns and characteristics
IEEE Network: The Magazine of Global Internetworking
Hi-index | 0.00 |
Real-time traffic aggregates are often characterized by time-varying nature. However, resources for this kind of traffic are usually allocated assuming busy hour stationary traffic characteristics. This assumption may lead to inefficient use of network resources when static resource reservation is used. In this paper we propose the algorithm for on-line estimation of the traffic state in terms of piecewise weakly stationary stochastic process. As a basic tool of the algorithm we use change-point statistical test allowing us to dynamically and automatically determine whether the traffic pattern changes and, if so, estimate new parameters of the traffic pattern. Assuming that a network may assign the required resources on-demand on a per-node basis, the proposed procedure helps to decide which amount of resources should be assigned to handle the traffic with given performance metrics. The proposed algorithm is well-suited for non-stationary behavior of the aggregated traffic where statistical parameters change in time. The computational complexity of the algorithm is low and it can be implemented on a traffic class basis making the proposal suitable for backbone routers.