Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!nntp.coast.net!dispatch.news.demon.net!demon!avondale.demon.co.uk!avondale.demon.co.uk!not-for-mail From: jfhall@avondale.demon.co.uk (John F Hall) Newsgroups: demon.ip.support,demon.tech.unix,comp.unix.bsd.freebsd.misc Subject: Re: Batch FTP and Web Pages Date: 14 Jul 1996 03:45:23 +0100 Organization: - Lines: 33 Message-ID: <4s9n03$179@avondale.demon.co.uk> References: <31D4AA3A.BC0@www.play-hookey.com> <4rr0us$fj@anorak.coverform.lan> <4rtrbh$2s8@avondale.demon.co.uk> <31E7146D.2FCE@www.play-hookey.com> NNTP-Posting-Host: avondale.demon.co.uk X-NNTP-Posting-Host: avondale.demon.co.uk In article <31E7146D.2FCE@www.play-hookey.com>, Ken Bigelow <kbigelow@www.play-hookey.com> wrote: >Just to check time lags, I did a reverse hop check to >demon-du.demon.co.uk, and came up with the following (at about 7:30 PM >EDT): > > 1. 206.161.179.129www.play-hookey.com (13 ms) > 2. 205.252.116.183asxxx.erols.com (159 ms) > 3. 205.252.116.164rtprime.erols.net (148 ms) > 4. 206.161.76.62 mae-east-h0/0.erols.net (178 ms) > 5. 192.41.177.245 mae-east.psi.net (171 ms) > 6. 38.1.3.1 ne.sc.psi.net (226 ms) > 7. 38.1.3.1 ne.sc.psi.net (244 ms) > 8. 204.6.105.2 <unknown> (418 ms) > 9. 194.159.252.98 ermin-router.router.demon.net (354 ms) >10. 158.152.1.222 demon-du.demon.co.uk (359 ms) > >I'm sure there are minor variations at different times of day and with >changes in demand, but this about says it. The problem of slow >connections is not within the US itself, and (according to a number of >posts in this excessively long and deep thread), not in the UK itself. >The bottleneck, as I and several others have said, is the bandwidth of >the <virtual> cable crossing the Atlantic. This being so, I'd like to >step back to the original topic of this thread... Eh? Those times don't show any queuing delays, they seem reasonable for the distances involved. You aren't forgetting that the Atlantic is *wide*, are you, and that packets will take time to cross? That's nothing to do with the rate at which packets can be transmitted. -- John F Hall jfhall@avondale.demon.co.uk Compuserve: 100016,1210