Automatic TCP send and receive socket buffer sizing

[王晓东]

My FTP is 202.119.123.7. From 202.119.117/24 to it, there are two routers.
traceroute to 202.119.123.7 (202.119.123.7), 64 hops max, 40 byte packets
 1  202.119.117.254 (202.119.117.254)  0.794 ms  0.809 ms  0.882 ms
 2  172.17.2.14 (172.17.2.14)  0.963 ms  0.736 ms  0.868 ms
 3  202.119.123.7 (202.119.123.7)  23.467 ms  23.499 ms  23.228 ms

Because the number of anonymous ftp users is unlimited, the FTP is very
busy. 
FreeBSD# w
 6:59PM  up 16 days,  5:12, 1 user, load averages: 0.84, 0.81, 0.76
USER             TTY      FROM              LOGIN@  IDLE WHAT
root             p0       202.119.117.45    6:54PM     - w

My question is why Debian have used the auto-sizing buffer, but FTP's
sending buffer for it is under FreeBSD and XP. My FreeBSD do not apply the
patch, it is 6.0-stable.

Thanks.

Wangxd

-----邮件原件-----
发件人: Andre Oppermann [mailto:andre at freebsd.org] 
发送时间: 2006年12月14日 18:50
收件人: 王晓东
抄送: freebsd-net at freebsd.org
主题: Re: Automatic TCP send and receive socket buffer sizing


The log lines prove that the automatic send buffer sizing was
working.  On local high speed networks with very low RTT (<1ms)
the effect is almost not noticeable.  On links with higher
bandwidth and larger RTT (>10ms Internet and Corporate WAN)
there can be a very large difference.  You can perform a test
by transferring the same large file twice, once with auto sizing
enabled and once without.  A good target is the European half
of FTP.FreeBSD.ORG [62.243.72.50] as it has a high bandwidth
connection and the send buffer auto scaling patch installed.

-- 
Andre