On the predictability of large transfer TCP throughput
Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications
IEEE Transactions on Parallel and Distributed Systems
On the predictability of large transfer TCP throughput
Computer Networks: The International Journal of Computer and Telecommunications Networking
An empirical study of bandwidth predictability in mobile computing
Proceedings of the third ACM international workshop on Wireless network testbeds, experimental evaluation and characterization
Application of Data Mining Algorithms to TCP throughput Prediction in HTTP Transactions
IEA/AIE '08 Proceedings of the 21st international conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems: New Frontiers in Applied Artificial Intelligence
Using Data Mining Algorithms in Web Performance Prediction
Cybernetics and Systems
An empirical evaluation of short-period prediction performance
SPECTS'09 Proceedings of the 12th international conference on Symposium on Performance Evaluation of Computer & Telecommunication Systems
Analysis of prediction performance of training-based models using real network traffic
International Journal of Computer Applications in Technology
MILCOM'09 Proceedings of the 28th IEEE conference on Military communications
Dynamic resource allocation DAMA alternatives study for satellite communications systems
MILCOM'09 Proceedings of the 28th IEEE conference on Military communications
Hi-index | 0.00 |
Distributed applications use predictions of network traffic to sustain their performance by adapting their behavior. The timescale of interest is application-dependent and thus it isnatural to ask how predictability depends on the resolution, or degree of smoothing, of the network traffic signal. To help answer this question we empirically study the one-step-ahead predictability, measured by the ratio of mean squared error to signal variance, of network traffic at different resolutions. A one-step-ahead prediction at a coarse resolution is a prediction ofthe average behavior over a long interval. We apply a wide range of linear and nonlinear time series models to a large number of packet traces, generating different resolution views of the traces through two methods: the simple binning approach used by several extant network measurement tools, and by wavelet-based approximations. The wavelet-based approach is a natural way to provide multiscale prediction to applications. We find that predictability seemsto be highly situational in practice-it varies widely from trace to trace. Unexpectedly, predictability does not always increase as the signal is smoothed. Half of the time there is a sweet spot at which the ratio is minimized and predictability is clearly the best. Also surprisingly, predictors that can capture non-stationarity and nonlinearity provide benefits only at very coarse resolutions.