Date: Wed, 9 Jul 2003 12:29:04 -0700 From: "Max Clark" <max.clark@media.net> To: "Dan Nelson" <dnelson@allantgroup.com> Cc: freebsd-hackers@freebsd.org Subject: RE: What ever happened with this? "eXperimental bandwidth delay product code" Message-ID: <ILENIMHFIPIBHJLCDEHKMEEKCJAA.max.clark@media.net> In-Reply-To: <20030709190214.GL39506@dan.emsphone.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> 6000000/8*.220 = 165Kbytes or 1.32Mbit/s I understand the BDP concept and the calculation to then generate the tcp window sizes. What I don't understand is this... How in the world is a windows 2000 box running commercial software able to push this link to 625KByte/s (5Mbit/s)???? How can I get similar results with FreeBSD? I don't care about any other traffic on the network at the same time as my transfer, just the raw performance of my transfer. Thanks, Max -----Original Message----- From: Dan Nelson [mailto:dnelson@allantgroup.com] Sent: Wednesday, July 09, 2003 12:02 PM To: Max Clark Cc: freebsd-hackers@freebsd.org Subject: Re: What ever happened with this? "eXperimental bandwidth delay product code" In the last episode (Jul 09), Max Clark said: > When you say it's got a specific purpose, I am looking for something > that will dynamically tune a 6Mbit/s, 220ms network link for bulk > (500MB) file transfers. Is this what I think it is, or should I be > looking at something else? Unless you're doing multiple simultaneous TCP connections it'll only slow you down. Your bw*delay product is 6000000/8*.220 = 165Kbytes, so telling ncftp to set its so-bufsize to say 200K, and telling your ftp daemon to do the same thing, should be all you need. -- Dan Nelson dnelson@allantgroup.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?ILENIMHFIPIBHJLCDEHKMEEKCJAA.max.clark>