[Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Stephen Hemminger]

This showed up on the end-to-end mailing list and might be of interest
to this group. It is interesting how many hosts are still using BIC
(probably RHEL/Centos 5). BIC is known to be broken and unfair.

From: Lisong Xu <lisongxu2@gmail.com>
Date: Tue, Mar 15, 2011 at 3:36 PM
Subject: [e2e] Identifying TCP congestion control algorithms, and
measurement results


Greetings,

We have recently developed a tool, called TCP Congestion Control
Avoidance Identification (CAAI), for actively identifying the TCP
congestion avoidance algorithm of a remote web server. We used CAAI to
measure the TCP algorithms of the top 5000 web sites in February 2011,
and got some preliminary results in which you might be interested.

# Only 16.85~25.58% of web servers still use the traditional AIMD.
# 14.36%, 15.82%, and 14.33% of web servers use BIC, CUBIC' (kernel
2.6.25 and before), and CUBIC (kernel 2.6.26 and after), respectively.
Total = 44.51%.
# 9.97% and 0.30~9.03% of web servers use CTCP' (Windows Server 2003
and XP Pro x64) and CTCP (Windows Server 2008, Vista, and 7),
respectively. Interestingly, CTCP' behaves very similar to HSTCP.
Total = 10.27~19%.
# Some web servers use non-default TCP algorithms (such as YEAH), some
web servers use some unknown TCP algorithms which are not available in
any major operating system family, and some web servers use abnormal
slow start algorithms.

More information is available at our project webpage
http://cse.unl.edu/~xu/research/TCPcensus.html.

Thanks
Lisong

--
Lisong Xu, Associate Professor
Computer Science & Engineering
University of Nebraska-Lincoln
http://cse.unl.edu/~xu

Re: [Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Jonathan Morton]

On 23 Mar, 2011, at 5:13 pm, Stephen Hemminger wrote:

> This showed up on the end-to-end mailing list and might be of interest
> to this group. It is interesting how many hosts are still using BIC
> (probably RHEL/Centos 5). BIC is known to be broken and unfair.

Interesting.  Is it possible to discover what TCP a specific host is using, eg. YouTube's main content servers?

 - Jonathan

Re: [Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Otto Solares]

On Wed, Mar 23, 2011 at 08:13:08AM -0700, Stephen Hemminger wrote:
> This showed up on the end-to-end mailing list and might be of interest
> to this group. It is interesting how many hosts are still using BIC
> (probably RHEL/Centos 5). BIC is known to be broken and unfair.

Too those "web servers use abnormal slow start algorithms" could be
related to sites using IW10 or more for the initial cwnd:

http://blog.benstrong.com/2010/11/google-and-microsoft-cheat-on-slow.html

IW10 seems to be part now of mainline Linux since 2.6.37:

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=356f039822b8d802138f7121c80d2a9286976dbd
-
 Otto

> From: Lisong Xu <lisongxu2@gmail.com>
> Date: Tue, Mar 15, 2011 at 3:36 PM
> Subject: [e2e] Identifying TCP congestion control algorithms, and
> measurement results
> 
> 
> Greetings,
> 
> We have recently developed a tool, called TCP Congestion Control
> Avoidance Identification (CAAI), for actively identifying the TCP
> congestion avoidance algorithm of a remote web server. We used CAAI to
> measure the TCP algorithms of the top 5000 web sites in February 2011,
> and got some preliminary results in which you might be interested.
> 
> # Only 16.85~25.58% of web servers still use the traditional AIMD.
> # 14.36%, 15.82%, and 14.33% of web servers use BIC, CUBIC' (kernel
> 2.6.25 and before), and CUBIC (kernel 2.6.26 and after), respectively.
> Total = 44.51%.
> # 9.97% and 0.30~9.03% of web servers use CTCP' (Windows Server 2003
> and XP Pro x64) and CTCP (Windows Server 2008, Vista, and 7),
> respectively. Interestingly, CTCP' behaves very similar to HSTCP.
> Total = 10.27~19%.
> # Some web servers use non-default TCP algorithms (such as YEAH), some
> web servers use some unknown TCP algorithms which are not available in
> any major operating system family, and some web servers use abnormal
> slow start algorithms.
> 
> More information is available at our project webpage
> http://cse.unl.edu/~xu/research/TCPcensus.html.
> 
> Thanks
> Lisong
> 
> --
> Lisong Xu, Associate Professor
> Computer Science & Engineering
> University of Nebraska-Lincoln
> http://cse.unl.edu/~xu

Re: [Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Steve Bauer]

On Wed, Mar 23, 2011 at 3:14 PM, Otto Solares <solca@guug.org> wrote:
> IW10 seems to be part now of mainline Linux since 2.6.37:
>
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=356f039822b8d802138f7121c80d2a9286976dbd

Note this patch only changed the initial *receive* window, not the
initial congestion window.

"This patch changes the default initial receive window to 10 mss
(defined constant). The default window is limited to the maximum
of 10*1460 and 2*mss (when mss > 1460).

[....]

However, an initial congestion window of 10 mss is useless unless a TCP
receiver advertises an initial receive window of at least 10 mss.
Fortunately, in the large-scale Internet experiments we found that most
widely used operating systems advertised large initial receive windows
of 64KB, allowing us to experiment with a wide range of initial
congestion windows. Linux systems were among the few exceptions that
advertised a small receive window of 6KB. The purpose of this patch is
to fix this shortcoming.
"

Re: [Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Aki Nyrhinen]

On Wed, 2011-03-23 at 13:14 -0600, Otto Solares wrote:
> IW10 seems to be part now of mainline Linux since 2.6.37:

that was just the initial receive window. the upcoming 2.6.39 is making
iw10 default on the sender side too.

	Aki

Re: [Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Otto Solares]

On Wed, Mar 23, 2011 at 04:36:58PM -0400, Steve Bauer wrote:
> On Wed, Mar 23, 2011 at 3:14 PM, Otto Solares <solca@guug.org> wrote:
> > IW10 seems to be part now of mainline Linux since 2.6.37:
> >
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=356f039822b8d802138f7121c80d2a9286976dbd
> 
> Note this patch only changed the initial *receive* window, not the
> initial congestion window.

Thanks for the correction! then let's wait for the real deal as it
seems the default (IW3) is too low for current networks.
-
 Otto

Re: [Bloat] Fwd: Identifying TCP congestion control algorithms, and measurement results

[Otto Solares]

On Wed, Mar 23, 2011 at 11:17:55PM +0200, Aki Nyrhinen wrote:
> On Wed, 2011-03-23 at 13:14 -0600, Otto Solares wrote:
> > IW10 seems to be part now of mainline Linux since 2.6.37:
> 
> that was just the initial receive window. the upcoming 2.6.39 is making
> iw10 default on the sender side too.

Then it seems we'll need to use the iproute initcwnd parameter until
2.6.39, thank you!
-
 Otto