nanog mailing list archives
Re: UUNET Routing issues
From: Rafi Sadowsky <rafi-nanog () meron openu ac il>
Date: Sat, 5 Oct 2002 18:32:21 +0300 (IDT)
## On 2002-10-04 23:50 +0200 Iljitsch van Beijnum typed: IvB> IvB> Obviously "some" packet loss and jitter are normal. But how much is IvB> normal? Even at a few tenths of a percent packet loss hurts TCP IvB> performance. The only way to keep jitter really low without dropping large IvB> numbers of packets is to severely overengineer the network. That costs IvB> money. So how much are customers prepared to pay to avoid jitter? There may be better ways to keep "reasonable" jitter but that depends on what is "really low" jitter - care to define numbers ? IvB> IvB> In any case, delays of 1000 ms aren't within any accepted definition of IvB> "normal". Ever used a satellite link ? Practical RTT("normal" - end to end including the local loops at both sides) starts at about 600msec
With these delays, high-bandwidth batch applications will
IvB> monopolize the links and interactive traffic suffers. I'm assuming TCP since you didn't state otherwise TCP extensions for "fat pipes"(such as window scaling and SACK) disabled (as both sides of the TCP connection need to have them) IIRC the maximum TCP(theoretical)session BW under these conditions Is less than 1Mb/sec (for 600msec RTT) For a reality check you may want to have look at the links under "Satellite links and performance" on <http://www.internet-2.org.il/documents.html> (yes the docs are a bit dated but the principles aren't)
20 ms worth of
IvB> buffer space with RED would keep those high-bandwidth applications in IvB> check and allow a reasonable degree of interactive traffic. Maybe a IvB> different buffer size would be better, but the 20 ms someone mentioned IvB> seems as good a starting point as anything else. IvB> IvB> -- Rafi
Current thread:
- Re: UUNET is not the Internet (and neither is AOL), (continued)
- Re: UUNET is not the Internet (and neither is AOL) Sean Donelan (Oct 07)
- RE: UUNET is not the Internet (and neither is AOL) Cleve Mickles (Oct 07)
- Re: UUNET is not the Internet (and neither is AOL) Jared Mauch (Oct 07)
- RE: UUNET is not the Internet (and neither is AOL) Ryan Fox (Oct 07)
- Re: UUNET is not the Internet (and neither is AOL) Dave Israel (Oct 07)
- Re: UUNET is not the Internet (and neither is AOL) Petri Helenius (Oct 07)
- Re: UUNET Routing issues Petri Helenius (Oct 04)
- Re: UUNET Routing issues Valdis . Kletnieks (Oct 04)
- Re: UUNET Routing issues Petri Helenius (Oct 04)
- Re: UUNET Routing issues Iljitsch van Beijnum (Oct 04)
- Re: UUNET Routing issues Rafi Sadowsky (Oct 05)
- Re: UUNET Routing issues Iljitsch van Beijnum (Oct 05)
- Re: UUNET Routing issues Marshall Eubanks (Oct 05)
- Re: UUNET Routing issues Petri Helenius (Oct 05)
- Re: UUNET Routing issues Stephen Sprunk (Oct 03)
- Re: UUNET Routing issues Scott Granados (Oct 06)