Empirical Evaluation of Upstream Throughput in a SIS Access Network.pdfVIP

Empirical Evaluation of Upstream Throughput in a DOCSIS Access Network ˇ Swapnil Bhatia, Radim Bartos, and Chaitanya Godsay Department of Computer Science University of New Hampshire Durham, NH 03824, USA Email: sbhatia, rbartos, cgodsay@ Abstract— We present empirical measurements of the up- shortcomings of the various performance enhancing features stream throughput of a DOCSIS1 1.1 link. In contrast to provided by the protocol. all previous simulation-based studies, our measurements have been obtained from actual cable-modems (CMs) and head-ends, II . AN OVERVIEW OF THE DOCSI S 1. 1 MAC LAYER both from two different vendors each. We have constructed an exhaustive database of measurements of a large subset of the A DOCSIS network uses the existing cable television infras- space of parameters affecting upstream throughput. Using a well- tructure to deliver data services to subscribers. The network, known non-parametric hypothesis test, we query this database owing to the structure of the pre-existing cable television for obtaining statistically robust answers to key questions about plant, forms a tree with the root connected to a Cable the effect of parameter changes on the throughput. Our results indicate that for a single CM scenario, packet concatenation is Modem Termination System (CMTS). Subscribers connect to most effective whereas piggybacking is effective and better than the network through a cable modem (CM) conn