|
From: | Marcus D. Leech |
Subject: | Re: [Discuss-gnuradio] Calculating the delay of TCP link. |
Date: | Tue, 30 Apr 2013 15:45:33 -0400 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.24) Gecko/20111108 Fedora/3.1.16-1.fc14 Thunderbird/3.1.16 |
On 04/30/2013 03:40 PM, Sajjad Safdar wrote:
Well, no, because you have to actually *measure* the delay. Any real IP network has a number of components, each of which introduces time-varying latencies. It's a meaningless question to ask "what is the delay of a typical TCP channel". Because, well, there's no such thing. Is your TCP channel across the room on a piece of wire? Does it go through switches/routers? How many of them? Are all the elements running at the same notional bit-rate? What is the current latency of packets from userland, through the kernel, until they "hit the wire"? That varies both with system architecture and current system loading factors. You could probably boil it all down to several different delay contributions, assign them variables and weighting, and come up with an equation. But without estimates for the likely ranges of each parameter, it won't tell you anything useful, compared to actually measuring the channel. Frequently. -- Marcus Leech Principal Investigator Shirleys Bay Radio Astronomy Consortium http://www.sbrac.org |
[Prev in Thread] | Current Thread | [Next in Thread] |