TCP/IP performance with random loss and bidirectional congestion

Publisher Website
Google Scholar
Abstract
With the growth in Internet access services over net- works with asymmetric links such as asymmetric digital subscriber line (ADSL) and cable-based access networks, it becomes crucial to evaluate the performance of TCP/IP over systems in which the bottleneck link speed on the reverse path (i.e., the path followed by acknowledgment) is considerably slower than that on the forward path (i.e., the path followed by data packets). In this paper, we pro- vide guidelines for designing network control mechanisms for sup- porting TCP/IP, the widely used Internet transport protocol, over such asymmetric networks. The key results underlying these guide- lines are as follows. We determine the throughput as a function of buffering, round-trip times, and normalized asymmetry (defined as the ratio of the transmission time of acknowledgment (ACK) in the reverse path to that of data packets in the forward path). We identify three modes of operation which are dependent on the forward buffer size and the normalized asymmetry, and determine the conditions under which the forward link is fully utilized. We also show that drop-from-front discarding of ACKs on the reverse link provides performance advantages over other drop mechanisms in use. Asymmetry increases TCP's already high sensitivity to random packet losses that occur on a time scale faster than the connec- tion round-trip time (e.g., caused by transient bursts in real-time traffic). We generalize the by-now well-known ("TCP-friendly") relation relating the square root of the random loss probability to obtained TCP throughput, originally derived considering only data path congestion. Specifically, random loss leads to significant throughput deterioration when the product of the loss probability, the normalized asymmetry and the square of the bandwidth delay product is large. Congestion in the reverse path adds considerably to TCP's un- fairness when multiple connections share the reverse bottleneck link. We show how such problems can be alleviated by per-con- nection buffer and bandwidth allocation on the reverse path.

This publication has 10 references indexed in Scilit: