TCP - IP White Papers
A Machine Learning Approach to TCP Throughput Prediction
Overview TCP throughput prediction is an important capability in wide area overlay and multi-homed networks where multiple paths may exist between data sources and receivers. This paper describes a new, lightweight method for TCP throughput prediction that can generate accurate forecasts for a broad range of file sizes and path conditions. The author's method is based on Support Vector Regression modeling that uses a combination of prior file transfers and measurements of simple path properties. This paper calibrates and evaluates the capabilities of the throughput predictor in an extensive set of lab-based experiments where ground truth can be established for path properties using highly accurate passive measurements.
| Publisher | Association for Computing Machinery | File Format | |
|---|---|---|---|
| Date Published | July 2007 | Downloads | 1 |
| Format | White Papers | ||
| Topics | |||
-
Did you find this white paper useful? 2 out of 6 users found this white paper useful
White Paper RSS
Download
Register
