Date: Fri, 23 Dec 1994 20:24:01 -0500 (EST) From: "Michael C. Newell" <mnewell@lupine.nsi.nasa.gov> To: "Daniel Stephens (CSC)" <stephens@mabuse.cas.usf.edu> Cc: freebsd-questions@freebsd.org Subject: Re: PPP and SLIP are amazingly slow. Message-ID: <Pine.SUN.3.91.941223201747.3930E-100000@lupine.nsi.nasa.gov> In-Reply-To: <Pine.BSD.3.91.941223203514.702B-100000@mabuse.cas.usf.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
I use CSLIP over V.fc [28.8 when they make it...] modems with the DTE rate set to 38.4Kbs. (I use 38.4 because that's the maximum speed my server ports run; my server is a Sun Sparc 2.) I see pretty much the same results on my 1.1.5.1 system. Actually I don't think those values are so bad; remember that ping packets are transmitted, the other end receives them, then the other end transmits a response; sort of a half-duplex operation. Try FTPing a large binary file - on my system I generally get a 24Kbs connection [I've only gotten 28.8 once, and it immediately negotiated down]. At that speed I see compressed binary transfers of about 2.7KB/s, and occasionally 3.0KB/s. I think the key is to try to keep data flowing - on an external modem you want the RD (or TD if you're sending) light to stay on solid; then you'll be getting more of an idea of the actual line bandwidth. Since FTP uses sliding windows you should get a much more accurate result than with ping. Mike On Fri, 23 Dec 1994, Daniel Stephens (CSC) wrote: > Date: Fri, 23 Dec 1994 20:44:31 -0800 (GMT-0800) > From: Daniel Stephens (CSC) <stephens@mabuse.cas.usf.edu> > To: freebsd-questions@freebsd.org > Subject: PPP and SLIP are amazingly slow. > > I'm interested in knowing what other people are getting for ping times > using 64k ICMP packets between slip client/server machines. Mine are as > follows: > > PPP 230ms > SLIP 198ms > > This seems awfully slow between two DX2/66 machines with 14.4 modems. > Everything I can find to read never seems to touch on the speed aspect of > either of these two protocols, so these times may be perfectly normal, > but I don't believe so. Quick (very poor) math shows that this isn't > even 5,000 bps. Something seems strange here. > Anyway, if anyone would like to share their ping times with me > and then, perhaps point out what they did to their > modems/sliplog/slattach to get imporved throughput I'd be a very happy > somethingorother. > > Dan > _______________________________________________________________________________ > Daniel Stephens | This | Somewhat versed in : > PC Networks/Open Use Labs | space | FreeBSD 1.1.5.1 Listserv V 5.5 > College of Arts and Sciences |intent-| Gopher MudLib HTTPD 1.3 (HTML+) > University of South Florida |ionally|----------------------------------------- > stephens@chuma.cas.usf.edu |left _ | snappy? No, but quite practical > > > Thanks, Mike +--------------------------------------+------------------------------------+ |Mike Newell | The opinions expressed herein are | |NASA Science Internet Network Systems | my own, and do not necessarily | |Sterling Software, Inc. | reflect those of the NSI program, | |MNewell@nsipo.nasa.gov | Sterling Software, NASA, or anyone | |+1-202-434-8954 | else. | +--------------------------------------+------------------------------------+
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.SUN.3.91.941223201747.3930E-100000>